P

umami · Production library run 2

43.5Library score
Pass rate94.5%
Simplicity50.6
Problems $$10.38

Implementers use the real production library (unagi-chan).

ImplementerPass rateSimplicityCostScore
DeepSeek V4.1 Flash · mini-SWE96.7%54.7$1.32
51.5
GLM 5.3 Flash · mini-SWE95.3%48.0$7.12
36.6
GPT-5.6 Luna · Codex91.5%48.8$1.94
42.2
ProblemDeepSeek V4.1 FlashGLM 5.3 FlashGPT-5.6 LunaMean

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