P

arazu · Production library run 2

38.0Library score
Pass rate71.9%
Simplicity65.6
Problems $$26.04

Implementers use the real production library (rapier).

ImplementerPass rateSimplicityCostScore
DeepSeek V4.1 Flash · mini-SWE77.1%62.9$3.75
38.6
GLM 5.3 Flash · mini-SWE66.7%64.0$16.92
35.8
GPT-5.6 Luna · Codex72.0%69.9$5.37
39.5
ProblemDeepSeek V4.1 FlashGLM 5.3 FlashGPT-5.6 LunaMean
lander-lidar49.356.344.450.0scene-contact46.320.113.526.6world-contact74.437.130.047.1aircraft-raycast44.544.646.545.2capsule-training12.60.20.24.3physics-tick9.51.928.513.3offset-strike53.670.363.162.3layered-laser78.278.572.876.5swept-cargo39.51.238.226.3compound-scanner27.157.553.546.0sensor-transit-ledger49.048.263.253.5articulated-creature0.60.362.921.3cable-replay38.40.00.012.8damping44.741.645.944.0one-way36.743.346.642.2material-braking37.131.744.037.6platformer-ccd32.941.243.839.3grab-replay55.256.317.042.8keyed-platform-drop56.623.541.740.6ccd-wall3.513.814.210.5motor-target9.346.414.723.5rope-cut50.974.784.069.9