MMLU-by-task - College Medicine — leaderboard
Metric: Accuracy (%). Source: huggingface.co. 1257 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | falcon-180B | 68.21 |
| 2 | Mistral-7B-v0.1 | 66.47 |
| 3 | StableBeluga2 | 65.32 |
| 4 | Llama 2 70B Chat | 64.74 |
| 5 | Llama 2 70B Base | 64.16 |
| 6 | Llama 2 70B Chat (HF) | 60.12 |
| 7 | Llama 2 70B Chat GPTQ | 60.12 |
| 8 | vicuna-13B-v1.5 | 57.23 |
| 9 | LLaMA-65B | 55.49 |
| 10 | LLaMA-30B | 54.34 |
| 11 | OpenHermes-13B | 53.76 |
| 12 | falcon-40B Instruct | 53.18 |
| 13 | internlm-20B Chat | 52.6 |
| 14 | MythoMax-L2-13B | 52.02 |
| 15 | vicuna-33B-v1.3 | 50.29 |
Interactive version: theaggregate.ai/benchmark?slug=mmlu-by-task-college-medicine · How the rankings work · Data refreshed daily, snapshot 2026-07-22.