heretical · Opus 5.5 run 3

47.8Library score
Pass rate85.4%
Simplicity63.9
Problems $$11.67
Library $$12.10
The library
heretical · designed by Opus 5.5 (mini-SWE)4,265 source lines
ImplementerPass rateSimplicityCostScore
DeepSeek V4.1 Flash · mini-SWE82.4%68.4$3.84
48.6
GLM 5.3 Flash · mini-SWE85.3%64.1$5.17
46.9
GPT-5.6 Luna · Codex88.6%59.4$2.66
47.8
ProblemDeepSeek V4.1 FlashGLM 5.3 FlashGPT-5.6 LunaMean

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