Clinical Decision Making on MedChain (overall) benchmark leaderboard · SOTA2 Research

Clinical Decision Making on MedChain (overall)

62.43Specialty Referral Accuracy (Lv1)ClinicalAgents

Evaluation Results

Method	Links
ClinicalAgents2026.03		62.43	35.43	48.2	48	59.76	52.59	51.07
MedChain-Agents2026.03		59.37	35.31	43.82	45.28	58.63	50.41	48.8
ReConcile2026.03		58.82	34.31	40.26	47.9	57.74	48.69	47.95
ReAct2026.03		58.03	32.12	39.61	45.43	58.2	48.35	46.96
ColaCare2026.03		57.98	32.73	39.94	44.41	56.23	49.2	46.75
Intern-S12026.03		57.83	34.06	33.17	41.92	48.1	39.64	42.45
DeepSeek-R12026.03		57.44	31.24	34.75	—	49.68	42.96	43.21
Qwen3-VL-235B2026.03		57.19	36.04	28.67	39.15	57.98	50.17	44.87
Few-shot + CoT2026.03		56.8	30.47	38.66	44.34	57.62	48.39	46.08
RAG2026.03		56.75	30.41	37.89	44.99	57.81	48.76	46.1
Llama-4-Maverick-17B2026.03		56.2	33.86	30.75	35.79	56.75	40.48	42.31
AutoGen2026.03		56.15	32.43	41.56	44.04	57.64	50.22	47.01
GPT-5.22026.03		54.67	30.45	38.7	42.76	57.54	47.16	45.21
MDAgents2026.03		54.42	28.87	40.72	45.82	57.59	48.25	45.94
MedAgents2026.03		53.44	29.21	39.61	46.76	54.42	49.38	45.47
MedGemma-4B2026.03		40.14	24.77	26.3	40.63	54.67	43.8	38.38
FineMedLM-o1-8B2026.03		36.88	19.48	21.4	—	47.16	41.57	33.3
HuaTuoGPT-o1-7B2026.03		36.09	20.37	15.37	—	54.42	46.47	34.54