HELM AFR - MMLU College Medicine: leaderboard
Metric: EM. Source: crfm.stanford.edu. 22 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | Gemini 2.0 Flash | 72.57 |
| 2 | Claude 3.7 Sonnet (20250219) | 67.89 |
| 3 | Gemini 2.0 Flash Lite | 66.16 |
| 4 | DeepSeek V3 | 63.37 |
| 5 | GPT-4o (2024-08-06) | 61.85 |
| 6 | Llama 3.1 70B Instruct | 53.44 |
| 7 | Llama 3.1 405B Instruct | 50.71 |
| 8 | GPT-4 Turbo | 47.08 |
| 9 | Llama 3.3 70B Instruct | 46.45 |
| 10 | Claude 3.5 Haiku (20241022) | 46.3 |
| 11 | Qwen 2.5 72B Instruct | 44.35 |
| 12 | Gemma 2 27B (IT) | 44.04 |
| 13 | GPT-3.5 Turbo (0125) | 37.94 |
| 14 | Llama 3.1 8B Instruct | 36.63 |
| 15 | Gemma 2 9B (IT) | 36.42 |
Interactive version: theaggregate.ai/benchmark?slug=helm-afr-mmlu-college-medicine · How It Works · Data refreshed daily, snapshot 2026-09-08.