MMLU-by-task - Moral Disputes — leaderboard
Metric: Accuracy (%). Source: huggingface.co. 1257 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | falcon-180B | 80.06 |
| 2 | Llama 2 70B Base | 77.75 |
| 3 | StableBeluga2 | 77.46 |
| 4 | Llama 2 70B Chat | 75.72 |
| 5 | LLaMA-65B | 73.7 |
| 6 | Llama 2 70B Chat (HF) | 71.68 |
| 7 | Llama 2 70B Chat GPTQ | 71.68 |
| 8 | Mistral-7B-v0.1 | 71.1 |
| 9 | vicuna-33B-v1.3 | 68.5 |
| 10 | LLaMA-30B | 67.05 |
| 11 | OpenHermes-13B | 64.74 |
| 12 | internlm-20B Chat | 64.74 |
| 13 | MythoMax-L2-13B | 62.43 |
| 14 | Llama 2 13B Chat Base | 61.27 |
| 15 | falcon-40B Instruct | 61.27 |
Interactive version: theaggregate.ai/benchmark?slug=mmlu-by-task-moral-disputes · How the rankings work · Data refreshed daily, snapshot 2026-07-22.