/ github.com / gke-labs / eval-runner

pr-395-4ea4529 Completed

← Back to Runs
Repository
https://github.com/codebot-robot/eval-runner.git
Commit
4ea4529
Model
google/gemini-3.5-flash-lite
Age
1h
Tasks
2/2 (2 ok, 0 fail)
Total Samples
16 (samples 1-8)
Mean Success Rate
0.56
Total Cost
$0.0189
Tokens Consumed
54863
Latency (mean of per-task p50 / p95)
2.42s / 2.94s

Eval Tasks

eval phase rate 95% CI (0-1 axis) latency (p50/p95) tokens cost n method log path
docs/readme_comprehension Completed 1.00 [0.68, 1.00]
2.44s / 2.65s 52878 $0.0161 8/8 (samples 1-8) wilson gs://gkelabs-evalrunner-dev-eval-runner/runs/4ea452974d0d3cfb5ae35ce61c730e2c976d04ce/docs/readme_comprehension/2026-10-27T17-43-13-00-00_readme-comprehension_PmjK7drVkvtEo93Kg7GCfa.json ↗
docs/readme_comprehension_blind Completed 0.12 [0.02, 0.47]
2.41s / 3.23s 1985 $0.0028 1/8 (samples 1-8) wilson gs://gkelabs-evalrunner-dev-eval-runner/runs/4ea452974d0d3cfb5ae35ce61c730e2c976d04ce/docs/readme_comprehension_blind/2026-10-27T17-43-49-00-00_readme-comprehension-blind_cRmgkYDom5MVJyoqZ3kcPL.json ↗