Text · Style Control · Hard Prompts
Arena · 50 models
| Rank | Model | Score |
|---|---|---|
| 1 | Gemini 4 Carbon (High) | 1,548.1746 |
| 2 | Gemini 4 Argon (High) | 1,544.3494 |
| 3 | Claude Opus 5.5 (High) | 1,534.7645 |
| 4 | Claude Opus 4.6 (High) | 1,530.6 |
| 5 | Claude Fable 5 (High) | 1,529.336 |
| 6 | Claude Opus 4.6 | 1,525.8045 |
| 7 | Claude Opus 4.7 (High) | 1,525.2642 |
| 8 | Muse Wonder (Max) | 1,524.1982 |
| 9 | Claude Fable 5.1 (Max) | 1,523.6935 |
| 10 | Claude Opus 4.7 | 1,518.8432 |
| 11 | Gemini 3.8 Flash (High) | 1,517.0327 |
| 12 | Kimi K3 (Max) | 1,516.8174 |
| 13 | Claude Opus 5 (High) | 1,516.3225 |
| 14 | Muse Spark 1.3 (Max) | 1,516.057 |
| 15 | Claude Opus 5 (Max) | 1,514.5498 |
| 16 | Muse Spark 1.2 (xHigh) | 1,513.4069 |
| 17 | Claude Opus 4.8 (High) | 1,513.3362 |
| 18 | GPT-5.6 Sol (xHigh) | 1,510.9813 |
| 19 | Muse Spark 1.1 | 1,510.5905 |
| 20 | Gemini 3.7 Flash (High) | 1,509.1568 |
| 21 | Gemini 3.1 Pro Preview | 1,507.6212 |
| 22 | GPT-6.1 Sol (Max) | 1,506.3691 |
| 23 | Muse Spark | 1,505.8939 |
| 24 | Claude Opus 4.8 | 1,504.9045 |
| 25 | Claude Sonnet 5.5 (xHigh) | 1,504.8781 |
| 26 | GLM-5.3 (Max) | 1,504.7221 |
| 27 | MiMo-V2.6-Pro | 1,504.3394 |
| 28 | Qwen3.8-Max | 1,503.5839 |
| 29 | Claude Sonnet 4.6 | 1,503.1184 |
| 30 | Gemini 3 Pro | 1,502.2757 |
| 31 | GLM-5.3-Flash | 1,501.782 |
| 32 | GPT-5.5 (High) | 1,501.296 |
| 33 | Gemini 3.6 Flash (High) | 1,501.2369 |
| 34 | Claude Opus 4.5 (High, 32K, 20251101) | 1,500.7325 |
| 35 | GPT-6 Astra (Max) | 1,499.8516 |
| 36 | GPT-5.5 | 1,498.9225 |
| 37 | DeepSeek-V4.1-Flash-20260910 (Max) | 1,498.7269 |
| 38 | Claude Opus 4.5 (20251101) | 1,498.0744 |
| 39 | GPT-5.2 Chat (20260210) | 1,496.6329 |
| 40 | GPT-5.4 (High) | 1,496.5242 |
| 41 | GLM-5.2 (Max) | 1,496.1631 |
| 42 | Gemini 3.5 Flash (Medium) | 1,495.2964 |
| 43 | Gemini 3.5 Flash (High) | 1,495.131 |
| 44 | Qwen3.7-Max-Preview | 1,494.703 |
| 45 | MiMo-V2.5-Pro | 1,494.1855 |
| 46 | Gemini 3 Flash | 1,492.5908 |
| 47 | Qwen3.5-Max-Preview | 1,491.1521 |
| 48 | GPT-5.5 Instant | 1,491.1043 |
| 49 | Claude Sonnet 5 (High) | 1,490.5464 |
| 50 | GPT-5.6 Terra (xHigh) | 1,489.5275 |