 Claude 3.5 SonnetAnthropic |
| — |
| 92.00 |
| — |
| Anthropic |
| — |
 Llama3.1-405B InstructFacebook AI研究实验室 | 4,050 | 89.00 | 88.60 | Facebook AI研究实验室 | — |
 DeepSeek V2.5DeepSeek-AI | 2,360 | 89.00 | — | DeepSeek-AI | — |
 Amazon Nova Pro亚马逊 | — | 89.00 | — | 亚马逊 | — |
 Grok 2xAI | 2,690 | 88.40 | — | xAI | — |
 Codestral 25.01MistralAI | — | 86.60 | 80.20 | MistralAI | — |
 GPT-4OpenAI | 1,750 | 85.40 | 83.50 | OpenAI | — |
| Amazon Nova Lite亚马逊 | — | 85.40 | — | 亚马逊 | — |
| Llama3-400B-Instruct-InTrainingFacebook AI研究实验室 | 4,000 | 84.10 | — | Facebook AI研究实验室 | — |
| DeepSeek-V3DeepSeek-AI | 6,810 | 82.60 | — | DeepSeek-AI | — |
| Amazon Nova Micro亚马逊 | — | 81.10 | — | 亚马逊 | — |
 C4AI Command A (202503)CohereAI | 1,110 | 80.00 | — | CohereAI | — |
| Grok-1.5xAI | — | 74.10 | — | xAI | — |
| DeepSeek-V2-236B-ChatDeepSeek-AI | 2,360 | 73.80 | 61.40 | DeepSeek-AI | — |
 Qwen2.5-Max阿里巴巴 | — | 73.20 | 80.60 | 阿里巴巴 | — |
 DBRX Instructdatabricks | 1,320 | 70.10 | — | databricks | — |
| DeepSeek-V3-BaseDeepSeek-AI | 6,810 | 65.20 | 75.40 | DeepSeek-AI | — |
| Grok-1xAI | 3,140 | 63.20 | — | xAI | — |
| Qwen1.5-110B阿里巴巴 | 1,100 | 52.40 | 58.10 | 阿里巴巴 | — |
| GPT-3.5OpenAI | 1,750 | 48.10 | 52.20 | OpenAI | — |
| Mixtral-8×22B-MoEMistralAI | 1,410 | 45.10 | 71.20 | MistralAI | — |
| DeepSeek-V2-236BDeepSeek-AI | 2,360 | 40.90 | 66.60 | DeepSeek-AI | — |
 PaLM-CoderGoogle Research | 5,400 | 35.90 | 47.00 | Google Research | — |
| CodexOpenAI | 1,750 | 28.81 | — | OpenAI | — |
| PaLMGoogle Research | 5,400 | 26.20 | 47.00 | Google Research | — |
