pyda · Kimi K3 run 3

50.0Library score
Pass rate91.2%
Simplicity60.5
Problems $$11.95
Library $$19.01
The library
pyda · designed by Kimi K3 (mini-SWE)5,967 source lines
ImplementerPass rateSimplicityCostScore
DeepSeek V4.1 Flash · mini-SWE92.2%61.5$3.47
52.7
GLM 5.3 Flash · mini-SWE91.5%64.0$5.92
49.7
GPT-5.6 Luna · Codex89.9%56.4$2.56
47.5
ProblemDeepSeek V4.1 FlashGLM 5.3 FlashGPT-5.6 LunaMean

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