Arena Text — Hard Prompts (No Style Control)
Chat & Writing
Raw Arena rating for hard Text Arena prompts without style-control adjustment. Higher ratings indicate stronger preference in blind pairwise user votes.
Top models (higher is better)
| Model | Score |
|---|---|
| Opus 4.6 | 1527 |
| Claude Opus 5 | 1526 |
| Claude Fable 5 | 1513 |
| Opus 4.7 | 1504 |
| GPT-5.5 | 1488 |
| Muse Spark 1.1 | 1486 |
| GPT-5.4 | 1486 |
| Gemini 3.6 Flash | 1486 |
| Gemini 3.5 Flash | 1486 |
| MiMo-V2.5-Pro | 1485 |
| Kimi K3 | 1484 |
| Gemini 3.1 Pro Preview | 1484 |
| Sonnet 4.6 | 1484 |
| Qwen3.7 Max Preview | 1482 |
| Gemini 3 Pro Preview | 1482 |
| Opus 4.8 | 1482 |
| ERNIE 5.1 | 1480 |
| GPT-5.6 Sol | 1476 |
| Opus 4.5 | 1475 |
| GLM-5.1 | 1475 |
| Muse Spark | 1474 |
| GLM-5.2 | 1471 |
| Kimi K2.6 | 1469 |
| GPT-5.6 Terra | 1469 |
| Grok 4.5 | 1467 |
| Qwen3.7-Plus | 1465 |
| Gemini 3 Flash Preview | 1465 |
| Sonnet 5 | 1464 |
| DeepSeek-V4-Pro | 1462 |
| Sonnet 4.5 | 1462 |
| MiMo-V2-Pro | 1458 |
| Gemini 2.5 Pro | 1455 |
| GPT-5.1 | 1454 |
| Kimi K2.5 | 1454 |
| GLM-5 | 1452 |
| GPT-5.6 Luna | 1451 |
| Inkling | 1451 |
| MiMo-V2.5 | 1450 |
| MiniMax M3 | 1450 |
| Qwen3 Max Preview | 1449 |