canon · GPT-6 Luna run 3
36.4Run score
Pass rate89.0%
Simplicity45.1
Problems $$0.07
GPT-6 Luna uses the real production library (pydantic).
ImplementerPass rateSimplicityCostScore
GPT-6 Luna · mini-SWE89.0%45.1$0.07
ProblemGPT-6 Luna
Click a problem to see how each solution's score is built.