MMLU-by-task - High School Geography — leaderboard
Metric: Accuracy (%). Source: huggingface.co. 1257 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | Llama 2 70B Base | 87.88 |
| 2 | StableBeluga2 | 87.37 |
| 3 | falcon-180B | 86.36 |
| 4 | Llama 2 70B Chat | 85.86 |
| 5 | Llama 2 70B Chat (HF) | 80.81 |
| 6 | LLaMA-65B | 80.3 |
| 7 | Llama 2 70B Chat GPTQ | 80.3 |
| 8 | Mistral-7B-v0.1 | 77.27 |
| 9 | vicuna-33B-v1.3 | 76.26 |
| 10 | internlm-20B Chat | 75.25 |
| 11 | falcon-40B Instruct | 74.24 |
| 12 | LLaMA-30B | 71.21 |
| 13 | vicuna-13B-v1.5 | 70.71 |
| 14 | Llama 2 13B Chat Base | 70.2 |
| 15 | WizardLM-13B-V1.2 | 69.19 |
Interactive version: theaggregate.ai/benchmark?slug=mmlu-by-task-high-school-geography · How the rankings work · Data refreshed daily, snapshot 2026-07-22.