MMLU-by-task - Management — leaderboard
Metric: Accuracy (%). Source: huggingface.co. 1257 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | Llama 2 70B Chat | 84.47 |
| 2 | falcon-180B | 84.47 |
| 3 | Llama 2 70B Base | 83.5 |
| 4 | LLaMA-65B | 82.52 |
| 5 | StableBeluga2 | 82.52 |
| 6 | Mistral-7B-v0.1 | 81.55 |
| 7 | Llama 2 70B Chat GPTQ | 81.55 |
| 8 | Llama 2 70B Chat (HF) | 80.58 |
| 9 | LLaMA-30B | 77.67 |
| 10 | vicuna-33B-v1.3 | 77.67 |
| 11 | falcon-40B Instruct | 75.73 |
| 12 | Llama 2 13B Chat Base | 73.79 |
| 13 | vicuna-13B-v1.5 | 72.82 |
| 14 | WizardCoder-Python-34B-V1.0 | 72.82 |
| 15 | internlm-20B Chat | 72.82 |
Interactive version: theaggregate.ai/benchmark?slug=mmlu-by-task-management · How the rankings work · Data refreshed daily, snapshot 2026-07-22.