List your company

Software Engineering Problem Solving on SWE-Bench hard problem subset

10.67Add. SolvedLeanEvolve
4.436.057.679.29Jun 2, 2026

Evaluation results

MethodLinks
LeanEvolveModel=Qwen-3.5-27B2026.0610.6760
LeanEvolveModel=Kimi-K2.52026.068.6770
LeanEvolveModel=GPT-5.22026.06870.67
LeanEvolveModel=Gemma-4-31B2026.065.3365.33
LeanEvolveModel=GLM-52026.064.6762.67