canon · DeepSeek V4 Pro run 3

40.2Library score
Pass rate97.0%
Simplicity42.8
Problems $$4.77
Library $$0.21
The library
canon · designed by DeepSeek V4 Pro (mini-SWE)1,335 source lines
ImplementerPass rateSimplicityCostScore
DeepSeek V4.1 Flash · mini-SWE95.8%43.7$2.40
39.5
GLM 5.3 Flash · mini-SWE97.3%43.8$1.40
41.5
GPT-5.6 Luna · Codex98.0%41.1$0.97
39.6
ProblemDeepSeek V4.1 FlashGLM 5.3 FlashGPT-5.6 LunaMean

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