umami · Opus 5.5 run 1
50.1Library score
Pass rate94.2%
Simplicity57.5
Problems $$7.14
Library $$8.77
The library
umami · designed by Opus 5.5 (mini-SWE)2,742 source lines
ImplementerPass rateSimplicityCostScore
DeepSeek V4.1 Flash · mini-SWE97.2%63.5$1.59
GLM 5.3 Flash · mini-SWE92.9%58.8$4.02
GPT-5.6 Luna · Codex92.5%50.4$1.53
ProblemDeepSeek V4.1 FlashGLM 5.3 FlashGPT-5.6 LunaMean
Click a problem to see how each solution's score is built.