Steerability
Arena · 50 models
| Rank | Model | Score |
|---|---|---|
| 1 | Gemini 4 Carbon (High) | 0.1884 |
| 2 | Claude Opus 5.5 (Max) | 0.1853 |
| 3 | Claude Fable 5.1 (Max) | 0.1514 |
| 4 | GPT-6.1 Terra (Max) | 0.1407 |
| 5 | GPT-6 Astra (Max) | 0.1387 |
| 6 | GPT-6.1 Sol (Max) | 0.1346 |
| 7 | Gemini 4 Argon (High) | 0.1339 |
| 8 | Claude Sonnet 5.5 (Max) | 0.1127 |
| 9 | Muse Wonder Full (Max) | 0.1089 |
| 10 | GPT-6 Sol (Max) | 0.0974 |
| 11 | Claude Fable 5 (High) | 0.0888 |
| 12 | GPT-5.6 Sol (xHigh) | 0.0641 |
| 13 | Kimi K3.1 Pro (Max) | 0.0633 |
| 14 | Claude Opus 5 (Max) | 0.0531 |
| 15 | GPT-6 Luna (Max) | 0.0421 |
| 16 | Claude Sonnet 5 (High) | 0.0412 |
| 17 | Muse Spark 1.3 (Max) | 0.037 |
| 18 | Claude Haiku 5.5 (High) | 0.0346 |
| 19 | Muse Wonder (Max) | 0.0309 |
| 20 | DeepSeek-V4.1-Flash-20261017 (Max) | 0.0251 |
| 21 | MiMo-V2.6-Flash | 0.0247 |
| 22 | GPT-5.6 Terra (xHigh) | 0.0082 |
| 23 | MiMo-V2.6-Pro | 0.0075 |
| 24 | GPT-6 Luna (Medium) | 0.0055 |
| 25 | Kimi K3 (Max) | 0.0036 |
| 26 | Grok 4.6 (xHigh) | 0.0022 |
| 27 | Hy4 preview | 0.0018 |
| 28 | Gemini 3.8 Flash (High) | -0.0008 |
| 29 | Qwen3.8-Max | -0.0015 |
| 30 | Mistral Large 4 | -0.0049 |
| 31 | GLM-5.2 (Max) | -0.0052 |
| 32 | Step 5 Preview | -0.0063 |
| 33 | GLM-5.3 (Max) | -0.0085 |
| 34 | DeepSeek-V4-Pro (High, 20260813) | -0.0087 |
| 35 | Grok 4.7 (xHigh) | -0.0106 |
| 36 | Hy3 | -0.024 |
| 37 | Qwen3.8-Flash-Next | -0.0289 |
| 38 | GPT-6 Luna (Low) | -0.0292 |
| 39 | GLM-5.3-Flash | -0.0413 |
| 40 | MiniMax-M3.1-Flash | -0.0469 |
| 41 | Qwen3.8-27B | -0.0473 |
| 42 | Gemini 3.7 Flash (High) | -0.0576 |
| 43 | Gemini 3.1 Pro Preview | -0.0709 |
| 44 | Beam 1.0 501B A23B | -0.0728 |
| 45 | MiniMax-M3 | -0.1048 |
| 46 | Qwen3.7-Plus | -0.1112 |
| 47 | Muse Spark 1.2 (xHigh) | -0.1129 |
| 48 | Qwen3.7-Max | -0.1159 |
| 49 | Spark-X2.6 | -0.1213 |
| 50 | Muse Spark 1.1 | -0.1382 |