Quality eval

This is a preview with sample runs. Sign up and your own calls get scored automatically.

Current quality

80 / 100

7-day trend

↗ +3

  • Auto quality scan 8/8 06:004.20openai/gpt-5-miniCompleted
  • golden dataset v35.00openai/gpt-5-mini2026-08-07 11:30Completed
  • tone & safety sweep4.60anthropic/claude-haiku-4-52026-08-06 18:02Completed
  • Auto quality scan 8/6 03:00openai/gpt-5-miniFailed
  • prompt v2 walkthrough3.80openai/gpt-5-mini2026-08-05 09:14Completed