/ github.com / codebot-robot / eval-runner

pr-394-7b89482 Completed

← Back to Runs
Repository
https://github.com/codebot-robot/eval-runner.git
Commit
7b89482
Model
google/gemini-3.5-flash-lite
Age
1h
Tasks
1/2 (1 ok, 1 fail)
Total Samples
8 (samples 1-8)
Mean Success Rate
0.25
Artifacts
gs://gkelabs-evalrunner-dev-eval-runner/runs/7b89482b390c99ad225f27440b1e5dab14b781cd/
Total Cost
$0.0021
Tokens Consumed
1692
Latency (mean of per-task p50 / p95)
2.60s / 5.17s

Eval Tasks

View: All Metrics Accuracy / Rate Latency (p50/p95) Tokens Cost ($ USD)
eval phase rate 95% CI (0-1 axis) latency (p50/p95) tokens cost n method log path
docs/readme_comprehension Failed — — — — — — — — —
Detail: step git clone failed with exit code 128: fatal: destination path 'repo' already exists and is not an empty directory.
docs/readme_comprehension_blind Completed 0.25 [0.07, 0.59]
2.60s / 5.17s 1692 $0.0021 2/8 (samples 1-8) wilson gs://gkelabs-evalrunner-dev-eval-runner/runs/7b89482b390c99ad225f27440b1e5dab14b781cd/docs/readme_comprehension_blind/2026-10-27T17-34-54-00-00_readme-comprehension-blind_H2K8PfR7BQrARE2bEK9Ydu.json ↗
artifacts: gs://gkelabs-evalrunner-dev-eval-runner eval-runner 369fe00d3cd14b7fcab13efa8fa2934429d4ae27