Open Medical LLM - MMLU Clinical Knowledge: leaderboard
MMLU Clinical Knowledge subset: clinical medicine MCQ testing diagnostic reasoning, treatment protocols, and patient management.
Metric: Accuracy (%). Source: huggingface.co. Saturation forecast: Estimated already saturated. 193 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | GPT-4 | 86.4 | #438 |
| 2 | Llama 3 8B Slerp Biomed Chat Chinese | 75.09 | #1215 |
| 3 | GPT-3.5 Turbo (1106) | 74.71 | #961 |
| 4 | Liberated-Qwen1.5-14B | 73.96 | #1097 |
| 5 | StopCarbon-10.7B-v5 | 73.58 | #1044 |
| 6 | Sakura-SOLAR-Instruct-CarbonVillain-en-10.7B-v2-slerp | 73.58 | #1062 |
| 7 | CarbonVillain-en-10.7B-v4 | 73.58 | #1058 |
| 8 | Daredevil-8B-abliterated-dpomix | 73.58 | #838 |
| 9 | NeuralLLaMa-3-8B-ORPO-v0.3 | 73.21 | #1174 |
| 10 | Daredevil-8B-abliterated | 72.83 | #846 |
| 11 | Master-Yi-9B | 72.83 | #1119 |
| 12 | suzume-llama-3-8B-multilingual | 72.45 | #1084 |
| 13 | Meta-Llama-3-8B-Instruct-abliterated-v3 | 72.45 | #1208 |
| 14 | NeuralLLaMa-3-8B-DT-v0.1 | 72.45 | #923 |
| 15 | Yi-1.5-9B-32K | 72.08 | #1412 |
No result here: #3 Claude Opus 5.5, #6 GPT-6 Astra, #7 GPT-6.1 Sol.
Interactive version: theaggregate.ai/benchmark?slug=open-medical-llm-mmlu-clinical-knowledge · How It Works · Data refreshed daily, snapshot 2026-10-20.