Text · Multi Turn
Arena · 50 models
| Rank | Model | Score |
|---|---|---|
| 1 | Gemini 4 Carbon (High) | 1,564.6757 |
| 2 | Gemini 4 Argon (High) | 1,541.7713 |
| 3 | Claude Opus 4.6 (High) | 1,512.1634 |
| 4 | Claude Opus 4.6 | 1,505.5736 |
| 5 | Claude Opus 4.7 (High) | 1,505.2635 |
| 6 | Muse Wonder (Max) | 1,505.0678 |
| 7 | Claude Opus 4.7 | 1,503.9836 |
| 8 | Gemini 3.8 Flash (High) | 1,503.6679 |
| 9 | Claude Fable 5 (High) | 1,503.4201 |
| 10 | Claude Opus 5.5 (High) | 1,500.2021 |
| 11 | Claude Opus 5 (Max) | 1,498.8192 |
| 12 | Muse Spark 1.2 (xHigh) | 1,496.1595 |
| 13 | Claude Fable 5.1 (Max) | 1,495.3761 |
| 14 | Claude Opus 5 (High) | 1,494.3237 |
| 15 | Qwen3.8-Max | 1,491.0275 |
| 16 | Kimi K3 (Max) | 1,489.573 |
| 17 | Muse Spark 1.3 (Max) | 1,487.8775 |
| 18 | Gemini 3.1 Pro Preview | 1,486.2639 |
| 19 | Gemini 3.7 Flash (High) | 1,484.862 |
| 20 | Gemini 3 Pro | 1,484.0487 |
| 21 | Muse Spark 1.1 | 1,483.7986 |
| 22 | GPT-5.4 (High) | 1,481.7923 |
| 23 | Gemini 3.6 Flash (High) | 1,481.2603 |
| 24 | Gemini 3.5 Flash (High) | 1,480.9541 |
| 25 | Qwen3.7-Max-Preview | 1,479.8588 |
| 26 | Qwen3.5-Max-Preview | 1,477.7503 |
| 27 | Muse Spark | 1,476.6979 |
| 28 | Gemini 3.5 Flash (Medium) | 1,475.9712 |
| 29 | GPT-5.5 (High) | 1,475.9202 |
| 30 | Claude Opus 4.8 (High) | 1,475.7005 |
| 31 | MiMo-V2.5-Pro | 1,475.5101 |
| 32 | GLM-5.1 | 1,472.3978 |
| 33 | Claude Opus 4.8 | 1,472.3672 |
| 34 | GPT-5.5 | 1,472.3471 |
| 35 | Gemini 3 Flash | 1,471.0894 |
| 36 | GLM-5.3-Flash | 1,471.052 |
| 37 | ERNIE 5.1 | 1,470.3442 |
| 38 | GLM-5.3 (Max) | 1,470.0073 |
| 39 | GLM-5.2 (Max) | 1,469.0457 |
| 40 | Claude Sonnet 5.5 (xHigh) | 1,468.1139 |
| 41 | DeepSeek-V4-Pro | 1,467.4295 |
| 42 | Claude Opus 4.5 (20251101) | 1,466.0948 |
| 43 | MiMo-V2.6-Pro | 1,465.551 |
| 44 | Claude Sonnet 4.5 (20250929) | 1,464.6355 |
| 45 | GPT-5.4 | 1,463.3168 |
| 46 | Claude Sonnet 4.6 | 1,463.2079 |
| 47 | GPT-5.6 Sol (xHigh) | 1,461.2232 |
| 48 | DeepSeek-V4.1-Flash-20260910 (Max) | 1,460.2626 |
| 49 | Claude Opus 4.5 (High, 32K, 20251101) | 1,460.2482 |
| 50 | Qwen3.7-Plus | 1,459.3399 |