加载中...
加载中...
本页面提供最新、最全面的大模型数学推理能力评测排行榜。我们通过 GSM8K、MATH、AIME 2025 等多个权威数学基准数据集，对包括 OpenAI 的 GPT-4o、Anthropic 的 Claude、阿里巴巴的 Qwen、DeepSeek-R1 等模型进行评测。
数据来源：DataLearnerAI
| 3 | GLM-4.6 | 98.60 | 0.00 | 0.00 | 0.00 |
| 4 | Kimi K2.5 | 96.10 | 0.00 | 0.00 | 0.00 |
| 5 | GLM-4.7 | 95.70 | 0.00 | 0.00 | 0.00 |
| 6 | DeepSeek V3.2 | 93.10 | 0.00 | 0.00 | 0.00 |
| 7 | o3-pro | 93.00 | 93.00 | 0.00 | 0.00 |
| 8 | Qwen3-235B-A22B-Thinking-2507 | 92.30 | 0.00 | 0.00 | 0.00 |
| 9 | DeepSeek-V3.1 Terminus | 90.00 | 0.00 | 0.00 | 0.00 |
| 10 | DeepSeek V3.2-Exp | 89.30 | 0.00 | 0.00 | 0.00 |
| 11 | DeepSeek-V3.1 | 88.40 | 93.10 | 0.00 | 0.00 |
| 12 | DeepSeek-R1-0528 | 87.50 | 91.40 | 98.00 | 0.00 |
| 13 | Intern-S1 | 86.00 | 0.00 | 0.00 | 0.00 |
| 14 | Gemini-2.5-Pro-Preview-05-06 | 83.00 | 92.00 | 98.80 | 0.00 |
| 15 | Step3 | 82.90 | 0.00 | 0.00 | 0.00 |
| 16 | Qwen3-235B-A22B | 81.50 | 85.70 | 98.00 | 96.40 |
| 17 | M2.1 | 81.00 | 0.00 | 0.00 | 0.00 |
| 18 | MiniMax M2 | 78.00 | 0.00 | 0.00 | 0.00 |
| 19 | Grok 3 | 77.10 | 84.20 | 0.00 | 0.00 |
| 20 | MiniMax-M1-80k | 76.90 | 86.00 | 96.80 | 0.00 |
| 21 | Claude Opus 4 | 75.50 | 76.00 | 98.20 | 0.00 |
| 22 | Kimi K2 0905 | 75.20 | 0.00 | 0.00 | 0.00 |
| 23 | MiniMax-M1-40k | 74.60 | 83.30 | 96.00 | 0.00 |
| 24 | Gemini 2.5 Flash | 72.00 | 88.00 | 0.00 | 0.00 |
| 25 | Qwen3-235B-A22B-2507 | 70.30 | 0.00 | 0.00 | 0.00 |
| 26 | DeepSeek-R1 | 70.00 | 79.80 | 97.30 | 0.00 |
| 27 | Magistral-Medium-2506 | 64.95 | 73.59 | 0.00 | 0.00 |
| 28 | Gemini 2.5 Flash-Lite | 63.10 | 0.00 | 0.00 | 0.00 |
| 29 | Claude Sonnet 3.7 | 54.80 | 23.30 | 82.20 | 0.00 |
| 30 | Kimi K2 | 54.00 | 69.60 | 97.40 | 0.00 |
| 31 | DeepSeek-V3-0324 | 47.70 | 59.40 | 94.00 | 96.30 |
| 32 | GPT-4.1 | 36.70 | 48.10 | 92.80 | 95.90 |
| 33 | ERNIE-4.5-VL-424B-A47B-Base | 35.10 | 0.00 | 0.00 | 0.00 |
| 34 | ERNIE-4.5-300B-A47B | 35.10 | 54.80 | 96.40 | 96.60 |
| 35 | Gemini 2.0 Flash Experimental | 29.70 | 0.00 | 0.00 | 0.00 |
| 36 | Kimi K2 Thinking | 100.00 | 0.00 | 0.00 | 0.00 |
| 37 | Kimi k1.5 (Short-CoT) | 0.00 | 0.00 | 94.60 | 0.00 |
| 38 | Qwen2.5-Max | 0.00 | 0.00 | 0.00 | 94.50 |
| 39 | Llama3.1-405B Instruct | 0.00 | 0.00 | 0.00 | 0.00 |
| 40 | Amazon Nova Pro | 0.00 | 0.00 | 0.00 | 0.00 |
| 41 | GLM-4.5 | 0.00 | 91.00 | 98.20 | 0.00 |
| 42 | GLM-4.5-Air | 0.00 | 89.40 | 98.10 | 0.00 |
| 43 | OpenAI o3-mini (high) | 0.00 | 87.00 | 97.90 | 0.00 |
| 44 | OpenAI o1 | 0.00 | 79.20 | 96.40 | 0.00 |
| 45 | Kimi k1.5 (Long-CoT) | 0.00 | 0.00 | 96.20 | 0.00 |
| 46 | Claude Sonnet 3.7-64K Extended Thinking | 0.00 | 80.00 | 96.20 | 0.00 |
| 47 | Llama 4 Behemoth Instruct | 0.00 | 0.00 | 95.00 | 0.00 |
| 48 | Grok 3 mini | 0.00 | 40.00 | 0.00 | 0.00 |
| 49 | GPT-4.5 | 0.00 | 36.70 | 90.70 | 0.00 |
| 50 | OpenAI o1-mini | 0.00 | 63.60 | 90.00 | 0.00 |
| 51 | DeepSeek-V3 | 0.00 | 39.00 | 87.80 | 0.00 |
| 52 | Grok-3 mini - Reasoning | 0.00 | 96.00 | 0.00 | 0.00 |
| 53 | Grok-3 - Reasoning Beta | 0.00 | 93.30 | 0.00 | 0.00 |
| 54 | GPT-4.1 mini | 0.00 | 49.60 | 0.00 | 0.00 |
| 55 | Grok 3.5 | 0.00 | 0.00 | 0.00 | 0.00 |
| 56 | GPT-4.1 nano | 0.00 | 29.40 | 0.00 | 0.00 |
| 57 | Gemini 2.0 Pro Experimental | 0.00 | 36.00 | 0.00 | 0.00 |