pr-391-b2-cand Completed
Eval Tasks
| eval | phase | rate | 95% CI (0-1 axis) | latency (p50/p95) | tokens | cost | n | method | log path | |
|---|---|---|---|---|---|---|---|---|---|---|
| docs/readme_comprehension_blind | Completed | 0.12 | [0.02, 0.47] | 2.40s / 3.61s | 1627 | $0.0020 | 1/8 (samples 9-16) | wilson | gs://gkelabs-evalrunner-dev-eval-runner/runs/184dfbc73f5a4c073be67666fc2bde04710a9e07/docs/readme_comprehension_blind/2026-10-27T15-47-11-00-00_readme-comprehension-blind_UMdSivNg9NXotZhqUo7mSV.json ↗ | |