SOTA Medical Reasoning on MedMCQA and PapersWithCode

86Accuracy

ReConcile

Updated 1d ago

Evaluation Results

Method	Links
ReConcile 2025.08		86
MedAgents 2025.08		85.7
TMA-AllCompon 2025.08		85.4
DyLAN 2025.08		84.6
MDAgents 2025.08		80.8
GPT-4.1 2025.07		80.58
GPT-4o 2025.07		76.17
Claude-3.5-Sonnet 2025.07		76
SMCS 2025.07		75.75
HuatuoGPT-o1-72B 2025.07		75.25
Open-source Upper Bound 2025.07		75.25
GPT-o3-mini 2025.07		74.92
Symbolic-MoE* 2025.07		74.88
Self-MoA 2025.07		74.88
Self Consistency (Best on Validation) 2025.07		74.83
Claude-3.7-Sonnet 2025.07		74.75
Majority Voting 2025.07		73.33
MoA 2025.07		73.08
DeepSeek-R1-Distill-Llama-70B 2025.07		72.5
Simple Router 2025.07		72.5
Llama-3.3-70B-Instruct 2025.07		72.25
Qwen3-32B 2025.07		72.17
Llama-3.3-Nemotron-Super-49B-v1 2025.07		70.92
GLM-Z1-32B-0414 2025.07		70.5
QwQ-32B 2025.07		69.83
Qwen-2.5-72B-Instruct 2025.07		69.08
DeepSeek-R1-Distill-Qwen-32B 2025.07		67.25
Gemma-3-27b-it 2025.07		64.58
TMA-AllCompon 2025.08		63.6
Qwen2.5-32b-Instruct 2025.07		62.92
Relational Aggregation 2026.05		62.83
Anchor (Refined Belief) 2026.05		61.12
GoAMax 2026.05		60.04
Anchor (Initial Belief) 2026.05		59.23
EXAONE-Deep-32B 2025.07		59.17
Majority (Refined Belief) 2026.05		58.87
GoAMean 2026.05		57.92
Majority (Initial Belief) 2026.05		57.47
Qwen2.5-Coder-32B-Instruct 2025.07		57.25
TeleChat2-35B-32K 2025.07		57.08
SC 2026.05		55.7
Self-MoA 2026.05		55.56
General 2026.05		55.22
Refine 2026.05		54.94
MoA 2026.05		54.94
ReConcile 2026.05		54.6
TMA-AllCompon 2025.08		53.3
Debate 2026.05		53.05
DyLAN 2025.08		52.4
ReConcile 2025.08		52
InternLM2.5-20B-Chat 2025.07		51.92
MedAgents 2025.08		50.8
Biomedical 2026.05		47
Code 2026.05		45.57
Math 2026.05		45.25
Finance 2026.05		42.08
Legal 2026.05		41.5
MDAgents 2025.08		38.5