–

uglypie · No library run 1

32.5Library score
Pass rate88.5%
Simplicity40.5
Problems $$2.63

Implementers get no library and write everything from scratch.

ImplementerPass rateSimplicityCostScore
DeepSeek V4.1 Flash · mini-SWE92.4%39.5$0.77
34.4
GLM 5.3 Flash · mini-SWE83.6%43.2$1.08
31.1
GPT-5.6 Luna · Codex89.4%38.7$0.78
31.9
ProblemDeepSeek V4.1 FlashGLM 5.3 FlashGPT-5.6 LunaMean

Click a problem to see how each solution's score is built.