Text · Hard Prompts
Arena · 50 models
| Rank | Model | Score |
|---|---|---|
| 1 | Gemini 4 Carbon (High) | 1,552.5086 |
| 2 | Gemini 4 Argon (High) | 1,541.7328 |
| 3 | Claude Opus 5.5 (High) | 1,536.0991 |
| 4 | Claude Fable 5.1 (Max) | 1,528.9959 |
| 5 | Claude Opus 5 (Max) | 1,526.4038 |
| 6 | Claude Opus 4.6 (High) | 1,524.6485 |
| 7 | Claude Opus 5 (High) | 1,523.3298 |
| 8 | Claude Opus 4.6 | 1,521.669 |
| 9 | MiMo-V2.6-Pro | 1,511.1768 |
| 10 | Gemini 3.8 Flash (High) | 1,508.282 |
| 11 | Claude Fable 5 (High) | 1,507.6001 |
| 12 | Muse Wonder (Max) | 1,506.9383 |
| 13 | Claude Opus 4.7 (High) | 1,505.6477 |
| 14 | Muse Spark 1.3 (Max) | 1,502.5362 |
| 15 | Claude Opus 4.7 | 1,499.5271 |
| 16 | Gemini 3.7 Flash (High) | 1,497.2772 |
| 17 | Qwen3.8-Max | 1,496.3632 |
| 18 | Kimi K3 (Max) | 1,495.5364 |
| 19 | Claude Sonnet 5.5 (xHigh) | 1,493.7649 |
| 20 | GLM-5.3-Flash | 1,492.2255 |
| 21 | Muse Spark 1.2 (xHigh) | 1,491.6424 |
| 22 | GLM-5.3 (Max) | 1,490.2866 |
| 23 | GPT-5.5 (High) | 1,488.9092 |
| 24 | Gemini 3.5 Flash (High) | 1,488.1939 |
| 25 | Muse Spark 1.1 | 1,485.8665 |
| 26 | MiMo-V2.5-Pro | 1,485.8056 |
| 27 | GPT-5.5 | 1,485.7869 |
| 28 | Gemini 3.6 Flash (High) | 1,485.6751 |
| 29 | Gemini 3.5 Flash (Medium) | 1,485.1641 |
| 30 | GPT-5.4 (High) | 1,484.9891 |
| 31 | Gemini 3.1 Pro Preview | 1,484.5649 |
| 32 | Claude Sonnet 4.6 | 1,484.3056 |
| 33 | GPT-5.6 Sol (xHigh) | 1,483.1449 |
| 34 | DeepSeek-V4.1-Flash-20260910 (Max) | 1,482.7904 |
| 35 | Qwen3.5-Max-Preview | 1,482.7424 |
| 36 | Claude Opus 4.8 (High) | 1,482.2205 |
| 37 | Qwen3.7-Max-Preview | 1,481.8714 |
| 38 | ERNIE 5.1 | 1,480.6365 |
| 39 | GLM-5.2 (Max) | 1,480.5632 |
| 40 | Gemini 3 Pro | 1,480.2621 |
| 41 | Claude Opus 4.5 (20251101) | 1,475.9836 |
| 42 | Muse Spark | 1,474.2018 |
| 43 | Claude Opus 4.8 | 1,474.1412 |
| 44 | MiMo-V2.6-Flash | 1,473.3554 |
| 45 | Claude Opus 4.5 (High, 32K, 20251101) | 1,473.2681 |
| 46 | GLM-5.1 | 1,473.0448 |
| 47 | Kimi K2.6 | 1,470.5389 |
| 48 | GPT-5.6 Terra (xHigh) | 1,468.9819 |
| 49 | GPT-5.4 | 1,468.6528 |
| 50 | GPT-6.1 Sol (Max) | 1,467.4611 |