umami · DeepSeek V4 Pro run 1
45.2Library score
Pass rate93.6%
Simplicity51.8
Problems $$7.02
Library $$0.24
The library
umami · designed by DeepSeek V4 Pro (mini-SWE)1,266 source lines
ImplementerPass rateSimplicityCostScore
DeepSeek V4.1 Flash · mini-SWE98.7%57.0$2.36
GLM 5.3 Flash · mini-SWE89.2%50.3$3.01
GPT-5.6 Luna · Codex92.6%48.2$1.65
ProblemDeepSeek V4.1 FlashGLM 5.3 FlashGPT-5.6 LunaMean
Click a problem to see how each solution's score is built.