MMLU-by-task - US Foreign Policy — leaderboard
Metric: Accuracy (%). Source: huggingface.co. 1257 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | Llama 2 70B Base | 92 |
| 2 | falcon-180B | 92 |
| 3 | StableBeluga2 | 89 |
| 4 | LLaMA-65B | 88 |
| 5 | Llama 2 70B Chat GPTQ | 88 |
| 6 | Llama 2 70B Chat (HF) | 87 |
| 7 | Mistral-7B-v0.1 | 86 |
| 8 | Llama 2 70B Chat | 86 |
| 9 | vicuna-33B-v1.3 | 84 |
| 10 | LLaMA-30B | 83 |
| 11 | falcon-40B Instruct | 83 |
| 12 | MythoMax-L2-13B | 82 |
| 13 | Llama 2 13B Chat Base | 81 |
| 14 | vicuna-13B-v1.5 | 81 |
| 15 | WizardLM-13B-V1.2 | 81 |
Interactive version: theaggregate.ai/benchmark?slug=mmlu-by-task-us-foreign-policy · How the rankings work · Data refreshed daily, snapshot 2026-07-22.