MATH
Updated Apr 3, 2026·1,462 views
- Problem Count
- 12500
- Institution
- Category
- 数学推理
- Metrics
- Accuracy
- Language
- 英语
- Difficulty
- Expert
Overview
一个具有挑战性的数学问题数据集，包含代数、微积分、几何、概率等多个领域。
Related resources
Latest MATH model rankings and full benchmark leaderboard
Browse the latest scores, model modes, release dates, and parameter sizes for MATH.
Source: DataLearnerAI
Data sourced primarily from official releases (GitHub, Hugging Face, papers), then benchmark leaderboards, then third-party evaluators. Learn about our data methodology
Model Mode Legend
MATH Rank
| Rank | Model | License | |||
|---|---|---|---|---|---|
 OpenAI o3-mini (high) Standard Mode | 97.90 | 2025-01-31 | Unknown | Closed | |
OpenAI o1 Standard Mode | 96.40 | 2024-12-05 | Unknown | Closed | |
 Hunyuan-A13B-Instruct Standard Mode | 94.30 | 2025-06-27 | 80B | Free Commercial | |
| 4 |  Gemini 2.0 Pro Experimental Standard Mode | 91.80 | 2025-02-05 | Unknown | Closed |
| 5 | Hunyuan-TurboS Standard Mode | 89.70 | 2025-03-10 | Unknown | Closed |
| 6 |  Gemma 3 - 27B (IT) Standard Mode | 89.00 | 2025-03-12 | 27B | Free Commercial |
| 7 |  DeepSeek-V3 Standard Mode | 87.80 | 2024-12-26 | 681B | Free Commercial |
| 8 | Gemini 2.0 Flash-Lite Standard Mode | 86.80 | 2025-02-05 | Unknown | Closed |
| 9 | Gemma 3 - 12B (IT) Standard Mode | 83.80 | 2025-03-12 | 12B | Free Commercial |
| 10 |  Qwen2.5-32B Standard Mode | 83.10 | 2024-09-18 | 32B | Free Commercial |
| 11 | Gemini 1.5 Pro Standard Mode | 82.90 | 2024-02-15 | Unknown | Closed |
| 12 |  Claude 3.5 Sonnet New Standard Mode | 78.30 | 2024-10-22 | Unknown | Closed |
| 13 |  Llama3.3-70B-Instruct Standard Mode | 77.00 | 2024-12-06 | 70B | Free Commercial |
| 14 |  Amazon Nova Pro Standard Mode | 76.60 | 2024-12-03 | Unknown | Closed |
| 15 |  Grok 2 Standard Mode | 76.10 | 2024-08-13 | 269B | Free Commercial |
| 16 | GPT-4o Standard Mode | 75.90 | 2024-05-13 | Unknown | Closed |
| 17 | Llama3.1-405B Instruct Standard Mode | 73.90 | 2024-07-23 | 405B | Free Commercial |
| 18 | Claude 3.5 Sonnet Standard Mode | 71.10 | 2024-06-21 | Unknown | Closed |
| 19 | GPT-4o mini Standard Mode | 70.20 | 2024-07-18 | Unknown | Closed |
| 20 |  Mistral-Small-3.2 Standard Mode | 69.42 | 2025-06-20 | 24B | Free Commercial |
| 21 |  C4AI Aya Vision 32B Standard Mode | 69.30 | 2025-03-04 | 32B | Non-Commercial |
| 22 | Mistral-Small-3.1-24B-Instruct-2503 Standard Mode | 69.30 | 2025-03-17 | 24B | Free Commercial |
| 23 | Claude 3.5 Haiku Standard Mode | 69.20 | 2024-10-22 | Unknown | Closed |
| 24 | Qwen2.5-Max Standard Mode | 68.50 | 2025-01-28 | Unknown | Closed |
| 25 | GPT-4o(2024-11-20) Standard Mode | 68.50 | 2024-11-20 | Unknown | Closed |
| 26 | Llama3.1-70B-Instruct Standard Mode | 67.80 | 2024-07-23 | 70B | Free Commercial |
| 27 |  Phi-4-mini-instruct (3.8B) Standard Mode | 64.00 | 2025-02-27 | 3.8B | Free Commercial |
| 28 | Qwen3-Next Standard Mode | 62.36 | 2025-09-11 | 80B | Free Commercial |
| 29 | Qwen2.5-72B Standard Mode | 62.10 | 2024-09-18 | 72.7B | Free Commercial |
| 30 | Llama 4 Maverick Standard Mode | 61.20 | 2025-04-05 | 400B | Free Commercial |
Scroll to load 12 more