roadkill · Kimi K3 run 1

15.0Library score
Pass rate67.4%
Simplicity34.1
Problems $$9.15
Library $$4.04
The library
roadkill · designed by Kimi K3 (mini-SWE)2,341 source lines
ImplementerPass rateSimplicityCostScore
DeepSeek V4.1 Flash · mini-SWE68.1%36.9$2.38
16.7
GLM 5.3 Flash · mini-SWE65.3%28.4$5.01
10.1
GPT-5.6 Luna · Codex68.4%35.9$1.76
18.1
ProblemDeepSeek V4.1 FlashGLM 5.3 FlashGPT-5.6 LunaMean

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