• Joined on 2026-07-22
qqtang pushed to main at qqtang/sskj 2026-07-29 09:51:50 +00:00
e53b2c7e4c fix: JSONL parsing in run_batch, add results/ to gitignore
qqtang pushed to main at qqtang/sskj 2026-07-29 09:51:41 +00:00
c839230c0b feat: add EAGLE speculative decoding experiment (dsv4_pro6000_sglang_tp16_eagle)
qqtang pushed to main at qqtang/sskj 2026-07-28 09:31:54 +00:00
6d3338244b fix: cuda-graph-max-bs=64 (256 OOMs during capture)
qqtang pushed to main at qqtang/sskj 2026-07-28 09:20:27 +00:00
cd362c5ca0 fix: max_running_requests=256, cuda_graph_max_bs=256 to match max concurrency 128
qqtang pushed to main at qqtang/sskj 2026-07-28 09:09:28 +00:00
da1d4d3d78 add tp = 16 deepseek v4 pro sglang bench
qqtang pushed to main at qqtang/sskj 2026-07-23 02:07:19 +00:00
d13f61f7b8 Add vLLM TP×DP matrix benchmarking scripts and Docker support