Open Medical LLM - MMLU Professional Medicine: leaderboard
MMLU Professional Medicine subset: board-exam-level questions testing professional medical knowledge and clinical decision-making.
Metric: Accuracy (%). Source: huggingface.co. Saturation forecast: Estimated already saturated. 193 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | GPT-4 | 93.8 | #438 |
| 2 | CarbonBeagle-11B-truthy | 75 | #1029 |
| 3 | Llama 3 8B Slerp Biomed Chat Chinese | 75 | #1215 |
| 4 | Llama 3 8B Instruct | 74.63 | #1129 |
| 5 | Daredevil-8B-abliterated | 74.63 | #846 |
| 6 | ChimeraLlama-3-8B-v3 | 74.63 | #1055 |
| 7 | Llama-3-8B-Instruct-abliterated-dpomix | 74.63 | #1049 |
| 8 | NeuralLLaMa-3-8B-ORPO-v0.3 | 74.26 | #1174 |
| 9 | Daredevil-8B-abliterated-dpomix | 74.26 | #838 |
| 10 | Barcenas-Llama3-8B-ORPO | 73.9 | #949 |
| 11 | NeuralLLaMa-3-8B-DT-v0.1 | 73.53 | #923 |
| 12 | SOLAR-10.7B-Instruct-v1.0 | 73.16 | #1075 |
| 13 | StopCarbon-10.7B-v5 | 73.16 | #1044 |
| 14 | Sakura-SOLAR-Instruct-CarbonVillain-en-10.7B-v2-slerp | 73.16 | #1062 |
| 15 | Llama 3 8B Ita | 72.79 | #926 |
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-professional-medicine · How It Works · Data refreshed daily, snapshot 2026-10-20.