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

base-2a0e7b7 Completed

← Back to Runs
Repository
https://github.com/gke-labs/eval-runner.git
Commit
2a0e7b7
Model
google/gemini-3.5-flash-lite
Age
3h
Tasks
2/2 (2 ok, 0 fail)
Total Samples
16 (samples 1-8)
Mean Success Rate
0.69
Total Cost
$0.0200
Tokens Consumed
59949
Latency (mean of per-task p50 / p95)
2.50s / 2.94s

Eval Tasks

eval phase rate 95% CI (0-1 axis) n method log path
docs/readme_comprehension Completed 1.00 [0.68, 1.00]
8/8 (samples 1-8) wilson gs://gkelabs-evalrunner-dev-eval-runner/runs/2a0e7b7d9f0d193c507add6fb23fe3925f8dfcac/docs/readme_comprehension/2026-10-27T17-31-09-00-00_readme-comprehension_jYvZnoicGToC3nvLmV9Yoa.json ↗
docs/readme_comprehension_blind Completed 0.38 [0.14, 0.69]
3/8 (samples 1-8) wilson gs://gkelabs-evalrunner-dev-eval-runner/runs/2a0e7b7d9f0d193c507add6fb23fe3925f8dfcac/docs/readme_comprehension_blind/2026-10-27T17-31-47-00-00_readme-comprehension-blind_8EqE8UUFVB8kjYzLxTMqGt.json ↗