Arena Text — Instruction Following (No Style Control)
Chat & Writing
Raw Arena rating for instruction-following Text Arena prompts without style-control adjustment. Higher ratings indicate stronger preference in blind pairwise user votes.
Top models (higher is better)
| Model | Score |
|---|---|
| Claude Opus 5.5 | 1539 |
| Gemini 4 Argon | 1528 |
| Opus 4.6 | 1521 |
| Claude Fable 5.1 | 1519 |
| Claude Opus 5 | 1518 |
| Claude Fable 5 | 1502 |
| Opus 4.7 | 1498 |
| Muse Wonder | 1493 |
| Gemini 3.8 Flash | 1492 |
| MiMo-V2.6-Pro | 1491 |
| Claude Sonnet 5.5 | 1491 |
| Gemini 3.7 Flash | 1486 |
| Kimi K3 | 1485 |
| GPT-5.6 Sol | 1483 |
| Qwen3.8-Max | 1481 |
| GLM-5.3 | 1480 |
| GLM-5.3 Flash | 1479 |
| Muse Spark 1.3 | 1479 |
| GPT-5.5 | 1478 |
| Opus 4.8 | 1476 |
| Sonnet 4.6 | 1476 |
| MiMo-V2.5-Pro | 1475 |
| DeepSeek V4.1 Flash | 1474 |
| Opus 4.5 | 1472 |
| GPT-5.4 | 1469 |
| Gemini 3.6 Flash | 1467 |
| GPT-6.1 Sol | 1467 |
| Gemini 3.5 Flash | 1467 |
| Gemini 3.1 Pro Preview | 1466 |
| GLM-5.2 | 1466 |
| Muse Spark 1.2 | 1463 |
| Qwen3.7 Max Preview | 1460 |
| ERNIE 5.1 | 1460 |
| Sonnet 4.5 | 1459 |
| Gemini 3 Pro Preview | 1458 |
| Muse Spark 1.1 | 1456 |
| GPT-5.6 Terra | 1455 |
| GPT-6 Astra | 1455 |
| GLM-5.1 | 1452 |
| Sonnet 5 | 1452 |