SOTA Clinical Reasoning on MedCaseReasoning and PapersWithCode

54.96Accuracy

GPT-5.0

Updated 7d ago

Evaluation Results

Method	Links
GPT-5.0 2026.05		54.96
DeepSeek-V4 2026.05		52.51
MedGuideX-9B 2026.05		41.36
Claude-Haiku-4.5 2026.05		40.02
RL with CPG-Derived Process Rewards 2026.05		37.68
MedGuideX-4B 2026.05		35.34
Qwen3.5-9B 2026.05		33.11
Qwen3.5-9B 2026.05		32.66
Qwen3.5-9B 2026.05		31.88
Hulu-Med-7B 2026.05		31.22
Fine-tuning with CPG 2026.05		28.87
Qwen3.5-4B 2026.05		28.65
Qwen3.5-4B 2026.05		28.65
Qwen3.5-4B 2026.05		28.54
MedReason-8B 2026.05		27.76
Lingshu-7B 2026.05		23.86
Kimi-2.6 2026.05		23.63
Llama-Aloe-Beta-8B 2026.05		21.18
HuatuoGPT-o1-8B 2026.05		19.29
BioMistral-7B 2026.05		17.28
MediPhi-Instruct-4B 2026.05		17.06
Llava-Med-v1.5-7B 2026.05		16.72
MedGemma-1.5-4B 2026.05		16.39
CPGPrompt 2026.05		14.16
Clinical-R1-3B 2026.05		13.6
MedAlpaca-7B 2026.05		12.26