Text · Style Control · Non-English
Arena · 50 models
| Rank | Model | Score |
|---|---|---|
| 1 | Gemini 4 Carbon (High) | 1,520.763 |
| 2 | Gemini 4 Argon (High) | 1,510.0306 |
| 3 | Claude Fable 5.1 (Max) | 1,497.3863 |
| 4 | Claude Opus 5.5 (High) | 1,497.0043 |
| 5 | Muse Wonder (Max) | 1,496.9151 |
| 6 | Claude Fable 5 (High) | 1,492.9015 |
| 7 | Claude Opus 4.7 (High) | 1,489.5536 |
| 8 | Claude Opus 4.6 (High) | 1,488.4724 |
| 9 | Gemini 3.8 Flash (High) | 1,486.9119 |
| 10 | Gemini 3.7 Flash (High) | 1,484.7815 |
| 11 | Muse Spark 1.2 (xHigh) | 1,484.6957 |
| 12 | Muse Spark 1.3 (Max) | 1,484.4685 |
| 13 | Claude Opus 4.6 | 1,483.0317 |
| 14 | Claude Opus 4.7 | 1,482.9604 |
| 15 | Muse Spark 1.1 | 1,482.9337 |
| 16 | Claude Opus 5 (High) | 1,481.9635 |
| 17 | Claude Opus 5 (Max) | 1,481.5803 |
| 18 | Gemini 3.1 Pro Preview | 1,479.0556 |
| 19 | Gemini 3 Pro | 1,477.3795 |
| 20 | Muse Spark | 1,477.1546 |
| 21 | Kimi K3 (Max) | 1,477.0979 |
| 22 | GPT-5.6 Sol (xHigh) | 1,475.2058 |
| 23 | GPT-5.5 (High) | 1,473.911 |
| 24 | Gemini 3.6 Flash (High) | 1,471.4099 |
| 25 | Qwen3.8-Max | 1,471.1806 |
| 26 | Claude Opus 4.8 (High) | 1,469.5811 |
| 27 | GPT-5.5 Instant | 1,469.3594 |
| 28 | Qwen3.7-Max-Preview | 1,469.1159 |
| 29 | Gemini 3.5 Flash (High) | 1,468.9806 |
| 30 | GPT-6.1 Sol (Max) | 1,468.347 |
| 31 | GPT-5.4 (High) | 1,468.1847 |
| 32 | Gemini 3.5 Flash (Medium) | 1,467.4384 |
| 33 | GPT-5.5 | 1,466.21 |
| 34 | GLM-5.3-Flash | 1,465.5842 |
| 35 | GLM-5.3 (Max) | 1,465.4468 |
| 36 | Gemini 3 Flash | 1,463.8006 |
| 37 | Claude Opus 4.8 | 1,463.7758 |
| 38 | GPT-5.2 Chat (20260210) | 1,463.1453 |
| 39 | Grok 4.20 Beta 1 | 1,463.1166 |
| 40 | GPT-6 Astra (Max) | 1,462.5946 |
| 41 | GLM-5.2 (Max) | 1,461.7612 |
| 42 | Grok 4.20 Beta (Reasoning, 0309) | 1,461.3736 |
| 43 | DeepSeek-V4.1-Flash-20260910 (Max) | 1,460.658 |
| 44 | MiMo-V2.6-Pro | 1,460.3757 |
| 45 | Grok 4.20 Multi-Agent Beta (0309) | 1,460.2452 |
| 46 | Claude Opus 4.5 (High, 32K, 20251101) | 1,459.9707 |
| 47 | Claude Sonnet 5.5 (xHigh) | 1,458.3694 |
| 48 | Claude Sonnet 4.6 | 1,457.1116 |
| 49 | Claude Opus 4.5 (20251101) | 1,456.724 |
| 50 | GPT-5.4 | 1,456.7102 |