MMLU-by-task - High School World History — leaderboard
Metric: Accuracy (%). Source: huggingface.co. 1257 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | Llama 2 70B Chat | 90.72 |
| 2 | falcon-180B | 89.87 |
| 3 | Llama 2 70B Base | 87.76 |
| 4 | StableBeluga2 | 86.08 |
| 5 | Llama 2 70B Chat (HF) | 84.39 |
| 6 | LLaMA-65B | 83.97 |
| 7 | Llama 2 70B Chat GPTQ | 83.97 |
| 8 | vicuna-33B-v1.3 | 82.7 |
| 9 | LLaMA-30B | 79.75 |
| 10 | internlm-20B Chat | 78.9 |
| 11 | Mistral-7B-v0.1 | 77.22 |
| 12 | vicuna-13B-v1.5 | 76.79 |
| 13 | MythoMax-L2-13B | 75.95 |
| 14 | trurl-2-13B-academic | 74.68 |
| 15 | OpenHermes-13B | 73.42 |
Interactive version: theaggregate.ai/benchmark?slug=mmlu-by-task-high-school-world-history · How the rankings work · Data refreshed daily, snapshot 2026-07-22.