MMLU-by-task - High School US History — leaderboard
Metric: Accuracy (%). Source: huggingface.co. 1257 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | Llama 2 70B Chat | 91.18 |
| 2 | Llama 2 70B Base | 89.71 |
| 3 | falcon-180B | 89.71 |
| 4 | StableBeluga2 | 88.73 |
| 5 | Llama 2 70B Chat GPTQ | 86.27 |
| 6 | Llama 2 70B Chat (HF) | 85.78 |
| 7 | LLaMA-65B | 83.82 |
| 8 | Mistral-7B-v0.1 | 79.9 |
| 9 | vicuna-33B-v1.3 | 78.92 |
| 10 | LLaMA-30B | 78.43 |
| 11 | OpenHermes-13B | 76.96 |
| 12 | MythoMax-L2-13B | 76.47 |
| 13 | internlm-20B Chat | 75.98 |
| 14 | Llama 2 13B Chat Base | 75 |
| 15 | WizardLM-13B-V1.2 | 75 |
Interactive version: theaggregate.ai/benchmark?slug=mmlu-by-task-high-school-us-history · How the rankings work · Data refreshed daily, snapshot 2026-07-22.