Text · Style Control · Instruction Following
Arena · 50 models
| Rank | Model | Score |
|---|---|---|
| 1 | Gemini 4 Carbon (High) | 1,524.6434 |
| 2 | Gemini 4 Argon (High) | 1,524.5324 |
| 3 | Claude Opus 5.5 (High) | 1,522.1671 |
| 4 | Claude Opus 4.6 (High) | 1,510.7638 |
| 5 | Claude Fable 5 (High) | 1,508.464 |
| 6 | Claude Opus 4.7 (High) | 1,502.2105 |
| 7 | Claude Fable 5.1 (Max) | 1,500.1792 |
| 8 | Muse Wonder (Max) | 1,497.9291 |
| 9 | Claude Opus 4.6 | 1,497.6751 |
| 10 | Claude Opus 5 (High) | 1,496.3652 |
| 11 | Claude Opus 5 (Max) | 1,494.5498 |
| 12 | Claude Opus 4.7 | 1,492.6496 |
| 13 | Claude Opus 4.8 (High) | 1,491.5694 |
| 14 | Gemini 3.8 Flash (High) | 1,491.3541 |
| 15 | GPT-5.6 Sol (xHigh) | 1,490.6805 |
| 16 | Kimi K3 (Max) | 1,489.9376 |
| 17 | Gemini 3.7 Flash (High) | 1,487.0431 |
| 18 | Claude Sonnet 5.5 (xHigh) | 1,486.4102 |
| 19 | GPT-6.1 Sol (Max) | 1,485.7147 |
| 20 | Claude Opus 4.5 (High, 32K, 20251101) | 1,483.5554 |
| 21 | Muse Spark 1.3 (Max) | 1,483.2242 |
| 22 | Gemini 3.1 Pro Preview | 1,480.2329 |
| 23 | GLM-5.3 (Max) | 1,479.1513 |
| 24 | Claude Opus 4.8 | 1,478.075 |
| 25 | GPT-5.5 (High) | 1,477.6173 |
| 26 | DeepSeek-V4.1-Flash-20260910 (Max) | 1,476.5007 |
| 27 | GPT-5.5 | 1,474.9359 |
| 28 | Qwen3.8-Max | 1,474.8235 |
| 29 | Muse Spark 1.2 (xHigh) | 1,474.6739 |
| 30 | Claude Sonnet 4.6 | 1,474.2749 |
| 31 | GLM-5.3-Flash | 1,474.2627 |
| 32 | Claude Opus 4.5 (20251101) | 1,474.2564 |
| 33 | Gemini 3.6 Flash (High) | 1,472.7914 |
| 34 | Gemini 3 Pro | 1,472.3539 |
| 35 | Muse Spark 1.1 | 1,472.3165 |
| 36 | GPT-6 Astra (Max) | 1,472.2463 |
| 37 | MiMo-V2.6-Pro | 1,472.0717 |
| 38 | GPT-5.4 (High) | 1,470.9085 |
| 39 | GLM-5.2 (Max) | 1,469.0063 |
| 40 | MiMo-V2.5-Pro | 1,467.4466 |
| 41 | Qwen3.5-Max-Preview | 1,466.8777 |
| 42 | Claude Sonnet 5 (High) | 1,466.5385 |
| 43 | Gemini 3.5 Flash (Medium) | 1,466.3042 |
| 44 | Gemini 3.5 Flash (High) | 1,465.2722 |
| 45 | Qwen3.7-Max-Preview | 1,465.1731 |
| 46 | Muse Spark | 1,464.4056 |
| 47 | Claude Sonnet 4.5 (High, 32K, 20250929) | 1,463.7922 |
| 48 | GPT-5.6 Terra (xHigh) | 1,462.5134 |
| 49 | Claude Sonnet 4.5 (20250929) | 1,461.9059 |
| 50 | GLM-5.1 | 1,461.808 |