Text · Polish
Arena · 50 models
| Rank | Model | Score |
|---|---|---|
| 1 | Gemini 4 Argon (High) | 1,537.5168 |
| 2 | Claude Fable 5.1 (Max) | 1,535.7577 |
| 3 | Claude Opus 5 (Max) | 1,521.2133 |
| 4 | Claude Opus 5 (High) | 1,520.889 |
| 5 | Gemini 3.5 Flash (High) | 1,517.1376 |
| 6 | Gemini 3.7 Flash (High) | 1,504.2524 |
| 7 | Gemini 3 Pro | 1,502.7129 |
| 8 | Gemini 3.8 Flash (High) | 1,499.4628 |
| 9 | Gemini 3.5 Flash (Medium) | 1,498.5062 |
| 10 | Claude Opus 4.6 (High) | 1,497.9628 |
| 11 | Muse Spark 1.1 | 1,497.2022 |
| 12 | Gemini 3.1 Pro Preview | 1,496.9342 |
| 13 | Gemini 3 Flash | 1,493.9822 |
| 14 | Claude Fable 5 (High) | 1,492.7302 |
| 15 | Muse Spark 1.3 (Max) | 1,491.4598 |
| 16 | Claude Opus 4.7 | 1,489.9538 |
| 17 | Claude Opus 4.6 | 1,488.5356 |
| 18 | Dola Seed 2.0 Pro | 1,486.9677 |
| 19 | ERNIE-5.0-Preview (1022) | 1,482.9185 |
| 20 | ERNIE-5.0-Preview (1203) | 1,482.7211 |
| 21 | Claude Opus 4.7 (High) | 1,479.0479 |
| 22 | Gemini 2.5 Pro | 1,478.9378 |
| 23 | Qwen3.8-Max | 1,475.3096 |
| 24 | Gemini 3.6 Flash (High) | 1,474.8036 |
| 25 | Muse Spark | 1,474.5531 |
| 26 | GPT-5.4 (High) | 1,472.4115 |
| 27 | Gemini 3 Flash (Minimal) | 1,468.4288 |
| 28 | ERNIE 5.1 | 1,467.8934 |
| 29 | ERNIE 5.0 (0110) | 1,467.1491 |
| 30 | GPT-5.5 (High) | 1,465.8302 |
| 31 | MiMo-V2.5-Pro | 1,463.455 |
| 32 | Kimi K3 (Max) | 1,462.1219 |
| 33 | GLM-5.3 (Max) | 1,461.6442 |
| 34 | DeepSeek-V4-Pro | 1,460.7475 |
| 35 | GPT-5.5 | 1,459.5678 |
| 36 | Claude Opus 4.8 | 1,458.3781 |
| 37 | DeepSeek-V4.1-Flash-20260910 (Max) | 1,457.6244 |
| 38 | Claude Opus 4.8 (High) | 1,457.1474 |
| 39 | GLM-5.2 (Max) | 1,456.6496 |
| 40 | Grok 4.1 (Thinking) | 1,454.9226 |
| 41 | GPT-5.4 | 1,454.7126 |
| 42 | DeepSeek-V4-Pro Preview (High) | 1,454.1317 |
| 43 | Qwen3.5-Max-Preview | 1,454.1282 |
| 44 | Grok 4.20 Multi-Agent Beta (0309) | 1,453.9901 |
| 45 | GPT-5.6 Sol (xHigh) | 1,452.9885 |
| 46 | GLM-5.1 | 1,451.8301 |
| 47 | GLM-5.3-Flash | 1,451.7704 |
| 48 | Grok 4.1 | 1,451.5936 |
| 49 | Inkling | 1,451.5127 |
| 50 | Grok 4.20 Beta (Reasoning, 0309) | 1,451.4406 |