MMLU-by-task - College Mathematics: leaderboard
Metric: Accuracy (%). Source: huggingface.co. 1257 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | Phi-1.5 | 41 |
| 2 | Llama 2 70B Chat | 41 |
| 3 | CodeLlama-34B-Instruct-hf | 41 |
| 4 | Mistral-7B-v0.1 | 40 |
| 5 | LLaMA-30B | 38 |
| 6 | Llama 2 70B Base | 37 |
| 7 | Llama 2 7B Chat | 36 |
| 8 | LLaMA-65B | 36 |
| 9 | LLaMA-13B | 36 |
| 10 | mpt-30B | 36 |
| 11 | vicuna-7B-v1.3 | 36 |
| 12 | Llama 2 7B Base | 35 |
| 13 | Llama 2 70B Chat (HF) | 34 |
| 14 | LLaMA-7B | 34 |
| 15 | GPT-NeoXT-Chat-Base-20B | 34 |
Interactive version: theaggregate.ai/benchmark?slug=mmlu-by-task-college-mathematics · How It Works · Data refreshed daily, snapshot 2026-09-05.