Open Medical LLM - MMLU College Medicine: leaderboard
MMLU College Medicine subset: college-level medical science questions spanning pathology, pharmacology, and clinical medicine.
Metric: Accuracy (%). Source: huggingface.co. Saturation forecast: Estimated already saturated. 193 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | GPT-4 | 76.9 | #438 |
| 2 | Llama 3 8B Slerp Biomed Chat Chinese | 68.79 | #1215 |
| 3 | Liberated-Qwen1.5-14B | 68.21 | #1097 |
| 4 | Yi-1.5-9B | 67.63 | #1241 |
| 5 | Yi-1.5-9B-32K | 67.63 | #1412 |
| 6 | Daredevil-8B-abliterated-dpomix | 67.05 | #838 |
| 7 | NeuralLLaMa-3-8B-DT-v0.1 | 66.47 | #923 |
| 8 | suzume-llama-3-8B-multilingual | 65.9 | #1084 |
| 9 | CarbonBeagle-11B-truthy | 65.9 | #1029 |
| 10 | Meta-Llama-3-8B-Instruct-abliterated-v3 | 65.9 | #1208 |
| 11 | Master-Yi-9B | 65.9 | #1119 |
| 12 | Llama-3-8B-Instruct-abliterated-dpomix | 65.9 | #1049 |
| 13 | Llama 3 8B Instruct | 65.32 | #1129 |
| 14 | Starling-LM-7B-beta | 65.32 | #1093 |
| 15 | SOLAR-10.7B-Instruct-v1.0 | 65.32 | #1075 |
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-college-medicine · How It Works · Data refreshed daily, snapshot 2026-10-20.