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 |
|---|---|
| Opus 4.6 | 1522 |
| Claude Opus 5 | 1515 |
| Claude Fable 5 | 1508 |
| Opus 4.7 | 1497 |
| Opus 4.8 | 1479 |
| GPT-5.5 | 1478 |
| Sonnet 4.6 | 1477 |
| MiMo-V2.5-Pro | 1475 |
| Kimi K3 | 1474 |
| GPT-5.6 Sol | 1472 |
| Opus 4.5 | 1472 |
| GPT-5.4 | 1471 |
| Gemini 3.6 Flash | 1470 |
| Gemini 3.1 Pro Preview | 1466 |
| Gemini 3.5 Flash | 1465 |
| Muse Spark 1.1 | 1461 |
| Qwen3.7 Max Preview | 1460 |
| Gemini 3 Pro Preview | 1459 |
| ERNIE 5.1 | 1459 |
| Sonnet 4.5 | 1459 |
| GLM-5.2 | 1455 |
| GLM-5.1 | 1454 |
| GPT-5.6 Terra | 1454 |
| Kimi K2.6 | 1449 |
| Sonnet 5 | 1448 |
| Grok 4.5 | 1447 |
| DeepSeek-V4-Pro | 1446 |
| Qwen3.7-Plus | 1444 |
| MiMo-V2-Pro | 1444 |
| Muse Spark | 1442 |
| GPT-5.6 Luna | 1441 |
| GPT-5.1 | 1440 |
| Gemini 3 Flash Preview | 1437 |
| Gemini 2.5 Pro | 1437 |
| MiniMax M3 | 1436 |
| Opus 4.1 | 1434 |
| MiMo-V2.5 | 1434 |
| Gemma 4 31B | 1431 |
| Kimi K2.5 | 1430 |
| GLM-5 | 1428 |