List your company

Software Engineering Problem Solving on SWE-Bench Verified

25.8Rounds TakenMini SWE Agent + LaMR
24.79231.59638.445.204May 14, 2026

Evaluation results

MethodLinks
Mini SWE Agent + LaMRBackbone Model=Claude Opus 4.6, Pruning Technique=LaMR2026.0525.8380760.40.4533
Mini SWE Agent + SWE-PrunerBackbone Model=Claude Opus 4.6, Pruning Technique=SWE-Pruner2026.0526.838376.610.4986.6
Mini SWE AgentBackbone Model=Claude Opus 4.6, Pruning Technique=None2026.0530.937875.6—0.467—
Mini SWE Agent + LaMRBackbone Model=Claude Sonnet 4.5, Pruning Technique=LaMR2026.0539.635971.81.20.63330.5
Mini SWE Agent + SWE-PrunerBackbone Model=Claude Sonnet 4.5, Pruning Technique=SWE-Pruner2026.0541.7360721.40.70123.1
Mini SWE AgentBackbone Model=Claude Sonnet 4.5, Pruning Technique=None2026.055135370.6—0.911—