Text · Factuality · Life & Physical & Social Science
Arena · 50 models
| Rank | Model | Score |
|---|---|---|
| 1 | Claude Fable 5.1 (Max) | 1,529.9293 |
| 2 | Gemini 4 Argon (High) | 1,528.4887 |
| 3 | Claude Opus 5 (Max) | 1,528.2543 |
| 4 | Claude Opus 5.5 (High) | 1,527.5627 |
| 5 | Claude Opus 4.6 (High) | 1,520.6425 |
| 6 | Muse Wonder (Max) | 1,517.5661 |
| 7 | Claude Opus 5 (High) | 1,513.6531 |
| 8 | Claude Fable 5 (High) | 1,513.3874 |
| 9 | Qwen3.5-Max-Preview | 1,512.5331 |
| 10 | Qwen3.8-Max | 1,511.9396 |
| 11 | Claude Opus 4.6 | 1,511.6808 |
| 12 | Gemini 3.7 Flash (High) | 1,509.7276 |
| 13 | Kimi K3 (Max) | 1,509.1111 |
| 14 | Gemini 3.8 Flash (High) | 1,508.6972 |
| 15 | GPT-5.5 | 1,506.8667 |
| 16 | GPT-5.5 (High) | 1,506.5463 |
| 17 | Claude Opus 4.7 (High) | 1,504.83 |
| 18 | Gemini 3.5 Flash (High) | 1,504.4325 |
| 19 | Muse Spark 1.3 (Max) | 1,502.64 |
| 20 | Claude Opus 4.7 | 1,500.0983 |
| 21 | Gemini 3 Pro | 1,499.8373 |
| 22 | GPT-5.4 (High) | 1,499.2172 |
| 23 | Muse Spark 1.2 (xHigh) | 1,498.6363 |
| 24 | GLM-5.3 (Max) | 1,498.5841 |
| 25 | Claude Opus 4.8 (High) | 1,497.1248 |
| 26 | Claude Sonnet 5.5 (xHigh) | 1,496.4966 |
| 27 | Gemini 3.1 Pro Preview | 1,495.6783 |
| 28 | Gemini 3.5 Flash (Medium) | 1,494.9271 |
| 29 | DeepSeek-V4.1-Flash-20260910 (Max) | 1,492.8752 |
| 30 | GLM-5.2 (Max) | 1,491.7587 |
| 31 | Gemini 2.5 Pro | 1,491.2962 |
| 32 | Gemini 3.6 Flash (High) | 1,491.2231 |
| 33 | MiMo-V2.5-Pro | 1,490.8473 |
| 34 | Claude Opus 4.8 | 1,489.5222 |
| 35 | GPT-6 Astra (Max) | 1,489.2837 |
| 36 | GPT-6.1 Sol (Max) | 1,489.2657 |
| 37 | Grok 4.5 | 1,488.8614 |
| 38 | GPT-5.4 | 1,488.8597 |
| 39 | GLM-5.3-Flash | 1,488.1406 |
| 40 | GPT-5.6 Sol (xHigh) | 1,487.9558 |
| 41 | Claude Sonnet 4.6 | 1,486.3111 |
| 42 | Qwen3-Max-Preview | 1,485.121 |
| 43 | GLM-5.1 | 1,485.0928 |
| 44 | GPT-5.6 Terra (xHigh) | 1,484.5962 |
| 45 | Muse Spark | 1,484.3666 |
| 46 | Gemini 3 Flash | 1,483.6039 |
| 47 | DeepSeek-V4-Pro | 1,482.8926 |
| 48 | Muse Spark 1.1 | 1,482.2134 |
| 49 | ERNIE 5.1 | 1,482.164 |
| 50 | DeepSeek-V4-Pro (High, 20260813) | 1,482.1555 |