List your company

Medical Question Answering on MedQA USMLE (Accuracy)

90.4AccuracyGemini-3-flash
25.24442.159559.07575.9905Apr 13, 2025May 14, 2025Jun 14, 2025Jul 15, 2025Aug 15, 2025Sep 15, 2025Oct 17, 2025

Evaluation results

MethodLinks
Gemini-3-flash2025.1090.4
CoT-SCModel=LLaMA3-70b, Quantization=int42025.0480
QM-ToTModel=LLaMA3-70b, Quantization=int42025.0479.7
CoT-AVGModel=LLaMA3-70b, Quantization=int42025.0478.65
ToTModel=LLaMA3-70b, Quantization=int42025.0476.72
QM-ToTModel=Qwen2.5-72b, Quantization=int42025.0474.25
ToTModel=Qwen2.5-72b, Quantization=int42025.0473
CoT-AVGModel=Qwen2.5-72b, Quantization=int42025.0472.37
AM-localNumber of Adapters=30, External Tools=false2025.1070.6
CoT-SCModel=Qwen2.5-72b, Quantization=int42025.0470.25
AM-fullNumber of Adapters=30, External Tools=true2025.1070.2
Qwen3.5-9BToken Budget=generous2025.1069.6
QM-ToTModel=LLaMA3.1-8b, Quantization=int42025.0469.49
ToTModel=LLaMA3.1-8b, Quantization=int42025.0460.77
CoT-SCModel=LLaMA3.1-8b, Quantization=int42025.0459.19
CoT-AVGModel=LLaMA3.1-8b, Quantization=int42025.0458.77
QM-ToTModel=LLaMA2-70b, Quantization=int42025.0450.25
ToTModel=LLaMA2-70b, Quantization=int42025.0435.25
CoT-AVGModel=LLaMA2-70b, Quantization=int42025.0434
Qwen3.5-9BToken Budget=strict2025.1028.4
CoT-SCModel=LLaMA2-70b, Quantization=int42025.0427.75