List your company

Medical Question Answering on MedQA-USMLE (letter exact-match)

79.8Accuracy (Letter Match)MAGE
63.88868.01972.1576.281May 11, 2026

Evaluation results

MethodLinks
MAGEBackbone=Llama-3.1-8B, Execution Scaffold=Opus, Number of seeds=52026.0579.8
MAGEBackbone=Llama-3.1-8B, Execution Scaffold=Sonnet, Number of seeds=52026.0574.6
MedpromptBackbone=Llama-3.1-8B2026.0574
0-shot CoTBackbone=Llama-3.1-8B, Evaluation Protocol=0-shot2026.0564.5