MMLU-by-task - Virology — leaderboard
Metric: Accuracy (%). Source: huggingface.co. 1257 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | falcon-180B | 56.02 |
| 2 | Mistral-7B-v0.1 | 55.42 |
| 3 | LLaMA-65B | 54.22 |
| 4 | Llama 2 70B Chat | 54.22 |
| 5 | StableBeluga2 | 53.61 |
| 6 | Llama 2 70B Base | 53.01 |
| 7 | Llama 2 70B Chat (HF) | 51.2 |
| 8 | vicuna-33B-v1.3 | 50 |
| 9 | Llama 2 70B Chat GPTQ | 50 |
| 10 | LLaMA-30B | 49.4 |
| 11 | falcon-40B Instruct | 49.4 |
| 12 | internlm-20B Chat | 48.8 |
| 13 | Llama 2 13B Chat Base | 48.19 |
| 14 | trurl-2-13B-academic | 48.19 |
| 15 | vicuna-13B-v1.5 | 47.59 |
Interactive version: theaggregate.ai/benchmark?slug=mmlu-by-task-virology · How the rankings work · Data refreshed daily, snapshot 2026-07-22.