2 Commits

Author SHA1 Message Date
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
sora
44303d8953 Allow resume when only sample limit changes, and keep CSV rows across runs.
Identity fingerprints ignore limit so later larger runs reuse cached predictions. Summary tables upsert by benchmark and take agent TTFT/latency from the report's per-request stats.

Co-authored-by: Cursor <cursoragent@cursor.com>
2026-09-03 03:54:26 +00:00