加载中...
加载中...
一个包含 974 个简单的 Python 编程问题的基准，用于评估模型生成代码的能力。
Source: DataLearnerAI
Data sourced primarily from official releases (GitHub, Hugging Face, papers), then benchmark leaderboards, then third-party evaluators. Learn about our data methodology
| 排名 | 模型 | |||
|---|---|---|---|---|
| 1 | Llama3.1-405B Instructdefault | 88.6 | 2024-07-23 | 4050 |
| 2 | Gemini 1.5 Prodefault | 87.8 | 2024-02-15 | 未知 |
| 3 | Llama3.3-70B-Instructdefault | 87.6 | 2024-12-06 | 700 |
| 4 | GPT-4o minidefault | 87.2 | 2024-07-18 | 未知 |
| 5 | Llama3.1-70B-Instructdefault | 86 | 2024-07-23 | 700 |
| 6 | Claude 3.5 Haikudefault | 85.6 | 2024-10-22 | 未知 |
| 7 | Qwen2.5-72Bdefault | 84.7 | 2024-09-18 | 727 |
| 8 | Qwen2.5-32Bdefault | 84 | 2024-09-18 | 320 |
| 9 | Hunyuan-A13B-Instructdefault | 83.86 | 2025-06-27 | 800 |
| 10 | Qwen2.5-Maxdefault | 80.6 | 2025-01-28 | 未知 |
| 11 | Codestral 25.01default | 80.2 | 2025-01-13 | 未知 |
| 12 | Codestraldefault | 78.2 | 2024-05-29 | 220 |
| 13 | Llama 4 Maverickdefault | 77.6 | 2025-04-05 | 4000 |
| 14 | Qwen2.5-7Bdefault | 74.9 | 2024-09-18 | 70 |
| 15 | 74.71 | 2025-03-17 | 240 | |
| 16 | Gemma 3 - 27B (IT)default | 74.4 | 2025-03-12 | 270 |
| 17 | C4AI Aya Vision 32Bdefault | 70.43 | 2025-03-04 | 320 |
| 18 | Llama3.1-8B-Instructdefault | 69.4 | 2024-07-23 | 80 |
| 19 | Llama 4 Scoutdefault | 67.8 | 2025-04-05 | 1090 |
| 20 | Phi-4-mini-instruct (3.8B)default | 65.3 | 2025-02-27 | 38 |
| 21 | Moonlight-16B-A3B-Instructdefault | 63.8 | 2025-02-23 | 160 |
| 22 | Gemma 2 - 9Bdefault | 62.2 | 2024-06-27 | 90 |
| 23 | Qwen2.5-3Bdefault | 57.1 | 2024-09-18 | 30 |
| 24 | Llama3.1-8Bdefault | 53.9 | 2024-07-23 | 80 |
| 25 | Mistral-7B-Instruct-v0.3default | 51.1 | 2024-05-22 | 70 |
| 26 | Llama-3.2-3Bdefault | 48.7 | 2024-09-18 | 32 |
| 27 | Amazon Nova Prodefault | 0 | 2024-12-03 | 未知 |