Open Medical LLM - MMLU College Biology: leaderboard
MMLU College Biology subset: college-level biology questions covering cell biology, genetics, ecology, and molecular biology.
Metric: Accuracy (%). Source: huggingface.co. Saturation forecast: Estimated already saturated. 193 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | GPT-4 | 95.1 | #438 |
| 2 | Barcenas-Llama3-8B-ORPO | 81.25 | #949 |
| 3 | llama-3-merged-linear | 81.25 | #887 |
| 4 | Llama 3 8B Ita | 80.56 | #926 |
| 5 | Master-Yi-9B | 80.56 | #1119 |
| 6 | NeuralLLaMa-3-8B-ORPO-v0.3 | 79.86 | #1174 |
| 7 | Yi-1.5-9B-32K | 79.17 | #1412 |
| 8 | suzume-llama-3-8B-multilingual | 79.17 | #1084 |
| 9 | Daredevil-8B-abliterated | 79.17 | #846 |
| 10 | Llama-3-SauerkrautLM-8B-Instruct | 79.17 | #895 |
| 11 | NeuralLLaMa-3-8B-DT-v0.1 | 79.17 | #923 |
| 12 | Llama-3-8B-Instruct-abliterated-v2 | 79.17 | #1136 |
| 13 | Myrrh_solar_10.7b_3.0 | 79.17 | #1042 |
| 14 | Yi-1.5-dolphin-9B | 79.17 | #1155 |
| 15 | Yi-1.5-9B | 78.47 | #1241 |
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-biology · How It Works · Data refreshed daily, snapshot 2026-10-20.