| 名次 | 模型 | 机构 | Elo | 来源 |
|---|---|---|---|---|
| 1 | Gemini 3.1 Pro Preview | 1,013 | — | |
| 2 | Gemini 3.5 Flash | 1,010 | — | |
| 3 | GPT-5.5 | OpenAI | 1,006 | — |
| 4 | Qwen3.6 Flash | Alibaba Cloud / Qwen Team | 1,006 | — |
| 5 | Claude Opus 4.8 (Fast) | Anthropic | 1,004 | — |
| 6 | MiniMax-M3 | MiniMax | 1,004 | — |
| 7 | Kimi K2.6 | Moonshot AI | 1,002 | — |
| 8 | Claude Fable 5 | Anthropic | 1,002 | — |
| 9 | Qwen3.5 Plus | Alibaba Cloud / Qwen Team | 1,002 | — |
| 10 | Qwen3.6 35B-A3B | Alibaba Cloud / Qwen Team | 1,002 | — |
| 11 | Seed-2.0-Mini | ByteDance | 1,000 | — |
| 12 | Seed-2.0-Lite | ByteDance | 1,000 | — |
| 13 | Claude Opus 4.7 (Fast) | Anthropic | 1,000 | — |
| 14 | Seed-2.0-Pro | ByteDance | 998 | — |
| 15 | Claude Opus 4.7 | Anthropic | 998 | — |
| 16 | Step 3.7 Flash | StepFun | 996 | — |
| 17 | Seed-2.1-Pro | ByteDance | 996 | — |
| 18 | Claude Opus 4.8 | Anthropic | 994 | — |
| 19 | Grok 4.3 | xAI | 986 | — |
| 20 | MiMo-V2.5 | Xiaomi | 981 | — |
更多基准见 全部评测基准;评测分数为事实数据,整理自公开评测体系。