Open Medical LLM - MMLU Anatomy: leaderboard
MMLU Anatomy subset: graduate-level anatomy multiple-choice questions testing knowledge of human anatomical structures and systems.
Metric: Accuracy (%). Source: huggingface.co. Saturation forecast: Estimated already saturated. 193 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | GPT-4 | 80 | #438 |
| 2 | Yi-1.5-9B | 68.89 | #1241 |
| 3 | Master-Yi-9B | 67.41 | #1119 |
| 4 | Yi-1.5-9B-32K | 65.93 | #1412 |
| 5 | suzume-llama-3-8B-multilingual | 65.93 | #1084 |
| 6 | Llama 3 8B Slerp Biomed Chat Chinese | 65.93 | #1215 |
| 7 | GPT-3.5 Turbo (1106) | 65.92 | #961 |
| 8 | Daredevil-8B-abliterated | 65.19 | #846 |
| 9 | ChimeraLlama-3-8B-v3 | 65.19 | #1055 |
| 10 | Daredevil-8B-abliterated-dpomix | 65.19 | #838 |
| 11 | Starling-LM-7B-beta | 64.44 | #1093 |
| 12 | Llama-3-8B-Instruct-abliterated-v2 | 64.44 | #1136 |
| 13 | Liberated-Qwen1.5-14B | 64.44 | #1097 |
| 14 | Llama 3 8B Instruct 262k | 63.7 | #1114 |
| 15 | Llama 3 8B Instruct | 62.22 | #1129 |
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-anatomy · How It Works · Data refreshed daily, snapshot 2026-10-20.