MMLU-by-task - Professional Accounting — leaderboard
Metric: Accuracy (%). Source: huggingface.co. 1257 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | Llama 2 70B Chat | 57.45 |
| 2 | falcon-180B | 57.09 |
| 3 | Llama 2 70B Base | 56.74 |
| 4 | StableBeluga2 | 51.77 |
| 5 | Llama 2 70B Chat (HF) | 50.71 |
| 6 | LLaMA-65B | 49.29 |
| 7 | Mistral-7B-v0.1 | 48.58 |
| 8 | OpenHermes-13B | 45.39 |
| 9 | Llama 2 70B Chat GPTQ | 44.33 |
| 10 | LLaMA-30B | 43.62 |
| 11 | internlm-20B Chat | 42.91 |
| 12 | vicuna-33B-v1.3 | 42.55 |
| 13 | vicuna-13B-v1.3 | 42.2 |
| 14 | WizardLM-13B-V1.2 | 42.2 |
| 15 | MythoMax-L2-13B | 40.07 |
Interactive version: theaggregate.ai/benchmark?slug=mmlu-by-task-professional-accounting · How the rankings work · Data refreshed daily, snapshot 2026-07-22.