uglypie · Kimi K3 run 3

58.9Library score
Pass rate90.0%
Simplicity69.9
Problems $$4.08
Library $$7.50
The library
uglypie · designed by Kimi K3 (mini-SWE)4,577 source lines
ImplementerPass rateSimplicityCostScore
DeepSeek V4.1 Flash · mini-SWE85.7%71.4$1.20
58.8
GLM 5.3 Flash · mini-SWE91.7%73.2$1.85
62.2
GPT-5.6 Luna · Codex92.5%65.0$1.02
55.6
ProblemDeepSeek V4.1 FlashGLM 5.3 FlashGPT-5.6 LunaMean

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