μ-MATH — leaderboard
Metric: Accuracy (%). Source: huggingface.co. 18 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | O1 | 89.5 |
| 2 | O1 Mini | 84.8 |
| 3 | QwQ 32B-Preview | 83.3 |
| 4 | DeepSeek R1 | 82.2 |
| 5 | Gemini 2.0 Flash (01-21) (Thinking) | 81.2 |
| 6 | Gemini 1.5 Pro | 80.7 |
| 7 | GPT-4o (2024-08-06) | 77.4 |
| 8 | Mistral Large 2 (Nov) Instruct (2411) | 76.7 |
| 9 | Qwen 2.5 72B Instruct | 75.7 |
| 10 | Claude 3.5 Sonnet | 75 |
| 11 | Gemini 1.5 Flash | 74.9 |
| 12 | Qwen2.5-Math-72B-Instruct | 74.4 |
| 13 | GPT-4o Mini (2024-07-18) | 72.5 |
| 14 | Qwen 2.5 7B Instruct | 69.9 |
| 15 | Qwen2.5-Math-7B-Instruct | 63.3 |
Interactive version: theaggregate.ai/benchmark?slug=math · How the rankings work · Data refreshed daily, snapshot 2026-07-22.