List your company
ResearchFormerly WizWand

Medical Question Answering on MedCalc-Bench

35.3AccuracyZero-Shot CoT
3.47611.7382028.262Nov 25, 2025
Updated 7mo ago

Evaluation results

MethodLinks
Zero-Shot CoTModel=Claude 3.7 Sonnet2025.1135.3
BFRSModel=o3 Mini2025.1134.7
MIPROv2Model=Claude 3.7 Sonnet2025.1134.7
MIPROv2Model=o3 Mini2025.1134.3
Zero-Shot CoTModel=o3 Mini2025.1134.2
BFRSModel=Claude 3.7 Sonnet2025.1134.1
HELM BaselineModel=o3 Mini2025.1134
Zero-Shot PredictModel=o3 Mini2025.1133.4
BFRSModel=GPT 4o2025.1127
MIPROv2Model=GPT 4o2025.1126.8
Zero-Shot CoTModel=GPT 4o2025.1126.6
Zero-Shot CoTModel=Gemini 2.0 Flash2025.1126.3
MIPROv2Model=Gemini 2.0 Flash2025.1125.4
BFRSModel=Gemini 2.0 Flash2025.1125.2
BFRSModel=Qwen3 4B2025.1122.7
Zero-Shot CoTModel=Llama 3.3 70B2025.1122.5
MIPROv2Model=Qwen3 4B2025.1121.5
HELM BaselineModel=Claude 3.7 Sonnet2025.1121
MIPROv2Model=Llama 3.3 70B2025.1121
Zero-Shot CoTModel=Qwen3 4B2025.1120.8
Zero-Shot PredictModel=Claude 3.7 Sonnet2025.1120.6
BFRSModel=Llama 3.3 70B2025.1120
HELM BaselineModel=GPT 4o2025.1118.8
Zero-Shot PredictModel=Gemini 2.0 Flash2025.1117
HELM BaselineModel=Gemini 2.0 Flash2025.1115.8
Zero-Shot PredictModel=GPT 4o2025.1115.7
Zero-Shot PredictModel=Qwen3 4B2025.1111.5
HELM BaselineModel=Llama 3.3 70B2025.1111.3
Zero-Shot PredictModel=Llama 3.3 70B2025.119.9
HELM BaselineModel=Qwen3 4B2025.114.7