sapi · Kimi K3 run 3

39.6Library score
Pass rate94.8%
Simplicity44.4
Problems $$2.77
Library $$5.27
The library
sapi · designed by Kimi K3 (mini-SWE)3,326 source lines
ImplementerPass rateSimplicityCostScore
DeepSeek V4.1 Flash · mini-SWE95.5%44.2$0.74
40.0
GLM 5.3 Flash · mini-SWE94.3%44.4$1.24
38.9
GPT-5.6 Luna · Codex94.7%44.7$0.79
39.9
ProblemDeepSeek V4.1 FlashGLM 5.3 FlashGPT-5.6 LunaMean

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