Skip to content
Hbench

Judy

judy

Judy's pass rate on every board it ran on. Scores on different boards don't add up, so this page doesn't add them.

On each board

Terminal-Bench 2.0

terminal-bench@2.0
Claude Opus 4.60.7.0, full, 89 tasks, 5 attempts, open
Ranks 1–6
8 harnesses
71.9%63.8–79.3Judy minus Terminus 2 2.0.0: +9.0 points, 95% interval +2.7 to +16.0, over 89 tasks both ran; a clear difference.

How Judy works

Each hexagon shows how Judy works, ranked among the harnesses on that board: outward is better.

71.9%pass rate, 95% interval 63.8–79.3

Judy 0.7.0, Lean contextJudy 0.7.0, Tests before finishingJudy 0.7.0, Direct pathJudy 0.7.0, No loopsJudy 0.7.0, Speed

Compare with another harness

Judy 0.7.0, Lean contextJudy 0.7.0, Tests before finishingJudy 0.7.0, Direct pathJudy 0.7.0, No loopsJudy 0.7.0, Speedmeta-harness 1.1.0, Lean contextmeta-harness 1.1.0, Cache usemeta-harness 1.1.0, Tests before finishingmeta-harness 1.1.0, Direct pathmeta-harness 1.1.0, No loopsmeta-harness 1.1.0, Speed
  • Judy 0.7.0 · 71.9%
  • meta-harness 1.1.0 · 76.4%