Atlas

Benchmarks

← All benchmarks

Arena Text — English (No Style Control)

Chat & Writing

Raw Arena rating for English-language Text Arena prompts without style-control adjustment. Higher ratings indicate stronger preference in blind pairwise user votes.

Top models (higher is better)

ModelScore
Gemini 4 Argon1526
Claude Opus 5.51522
Opus 4.61511
Claude Opus 51508
Claude Fable 5.11508
Muse Wonder1506
MiMo-V2.6-Pro1504
Gemini 3.8 Flash1498
Claude Fable 51498
Opus 4.71496
Muse Spark 1.31496
Qwen3.8-Max1492
Claude Sonnet 5.51486
Gemini 3.7 Flash1485
Muse Spark 1.21483
GLM-5.31483
Gemini 3.6 Flash1482
Kimi K31481
Gemini 3.5 Flash1480
Muse Spark1480
Gemini 3.1 Pro Preview1478
Gemini 3 Pro Preview1478
Muse Spark 1.11478
ERNIE 5.11477
MiMo-V2.5-Pro1477
GLM-5.11475
GLM-5.21475
DeepSeek V4.1 Flash1475
MiMo-V2.6-Flash1474
GLM-5.3 Flash1474
Sonnet 4.61472
GPT-5.51471
Qwen3.7 Max Preview1470
Opus 4.81469
Gemini 3 Flash Preview1467
GPT-5.41467
Opus 4.51461
Kimi K2.61460
DeepSeek-V4-Pro1459
Qwen3.7-Plus (source unspecified)1459
Loading Atlas data…