HELM MMLU - Virology: leaderboard
Metric: EM. Source: crfm.stanford.edu. 79 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | Claude 3.5 Sonnet (20240620) | 60.24 |
| 2 | GPT-4 Turbo | 60.24 |
| 3 | DeepSeek V3 | 59.64 |
| 4 | GPT-4o (2024-05-13) | 59.64 |
| 5 | GPT-4 (0613) | 59.64 |
| 6 | Mixtral 8x22B | 59.64 |
| 7 | Llama 3 70B | 59.04 |
| 8 | Mistral Large 2 (Jul) | 59.04 |
| 9 | Nova Pro | 59.04 |
| 10 | GPT-4 Turbo (Preview) | 59.04 |
| 11 | Yi Large (Preview) | 59.04 |
| 12 | Claude 3.5 Sonnet (20241022) | 58.43 |
| 13 | Qwen 2.5 72B Instruct | 58.43 |
| 14 | Llama 3.2 90B Vision Instruct | 58.43 |
| 15 | Qwen 1.5 72B | 58.43 |
Interactive version: theaggregate.ai/benchmark?slug=helm-mmlu-virology · How It Works · Data refreshed daily, snapshot 2026-09-08.