sapi · Kimi K3 run 1

38.4Library score
Pass rate94.6%
Simplicity42.5
Problems $$2.60
Library $$2.93
The library
sapi · designed by Kimi K3 (mini-SWE)2,450 source lines
ImplementerPass rateSimplicityCostScore
DeepSeek V4.1 Flash · mini-SWE96.0%41.1$0.66
37.5
GLM 5.3 Flash · mini-SWE90.4%43.9$1.10
36.8
GPT-5.6 Luna · Codex97.5%42.7$0.83
40.9
ProblemDeepSeek V4.1 FlashGLM 5.3 FlashGPT-5.6 LunaMean

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