sora 8b0a413cf1 Drop Harbor trial wall-clock from summary perf metrics. ...

Agent CSV rows now use jsonl per-call stats or the report request summary only, so TTFT/latency stay on the same request口径.

Co-authored-by: Cursor <cursoragent@cursor.com>

2026-09-03 04:11:23 +00:00

..

case

fix: terminal_bench_v2_1 result collection, apt mirror, and restore run.py env var

2026-08-05 03:56:44 +00:00

fingerprint

stop tracking fingerprint markdown docs (keep files on disk)

2026-09-03 02:38:35 +00:00

images_load

Preload Pier uv and mini-swe-agent wheels so DeepSWE sandbox builds skip GitHub.

2026-09-02 09:44:17 +00:00

tests

Drop Harbor trial wall-clock from summary perf metrics.

2026-09-03 04:11:23 +00:00

cleanup_docker_images.sh

feat: add docker image cleanup script for benchmark images

2026-08-05 05:38:20 +00:00

collect_results.py

Drop Harbor trial wall-clock from summary perf metrics.

2026-09-03 04:11:23 +00:00

diagnose_swe_images.py

Optimize SWE-bench build_images: skip make_test_spec when images are pre-loaded

2026-07-27 08:46:15 +00:00

load_swe_images.sh

update benchmark suites and evaluation docs

2026-08-18 03:19:01 +00:00

perf_backup.py

chore: update evalscope runner, thinking config and webui

2026-07-31 02:36:51 +00:00

pull_swe_bench_images.py

feat(pull_swe_bench_images): support comma-separated --dataset

2026-07-23 02:34:49 +00:00

run.py

Allow resume when only sample limit changes, and keep CSV rows across runs.

2026-09-03 03:54:26 +00:00

save_swe_images.sh

Add SWE image save/load scripts and update Docker config docs

2026-07-24 06:27:49 +00:00

set.sh

docs: add resource prep section and sync docker/bash helpers

2026-07-23 02:18:15 +00:00