Arena Text — Hard Prompts English (No Style Control)
Chat & Writing
Raw Arena rating for hard English 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 | 1532 |
| Claude Opus 5 | 1524 |
| Claude Fable 5 | 1511 |
| Opus 4.7 | 1504 |
| MiMo-V2.5-Pro | 1497 |
| Sonnet 4.6 | 1494 |
| Gemini 3.6 Flash | 1488 |
| Kimi K3 | 1488 |
| Muse Spark 1.1 | 1484 |
| GLM-5.1 | 1484 |
| ERNIE 5.1 | 1484 |
| GPT-5.5 | 1482 |
| Opus 4.8 | 1482 |
| Gemini 3.1 Pro Preview | 1481 |
| Opus 4.5 | 1481 |
| GPT-5.4 | 1480 |
| Gemini 3.5 Flash | 1479 |
| Gemini 3 Pro Preview | 1478 |
| Muse Spark | 1476 |
| Qwen3.7 Max Preview | 1474 |
| GPT-5.6 Sol | 1474 |
| MiMo-V2-Pro | 1473 |
| Sonnet 4.5 | 1471 |
| GLM-5.2 | 1470 |
| Kimi K2.6 | 1469 |
| Sonnet 5 | 1466 |
| Grok 4.5 | 1465 |
| Qwen3.7-Plus | 1465 |
| DeepSeek-V4-Pro | 1464 |
| MiMo-V2.5 | 1461 |
| MiniMax M3 | 1461 |
| GPT-5.6 Luna | 1460 |
| GLM-5 | 1459 |
| Kimi K2.5 | 1459 |
| GPT-5.6 Terra | 1458 |
| Gemini 3 Flash Preview | 1458 |
| GPT-5.1 | 1458 |
| GLM 5V Turbo | 1456 |
| Qwen3.6 Plus (2026-04-02) | 1455 |
| Gemma 4 31B | 1454 |