Atlas

Benchmarks

← All benchmarks

Arena Text — Writing, Literature & Language (No Style Control)

Professional Work

Raw Arena rating for writing, literature, and language industry prompts without style-control adjustment. Higher ratings indicate stronger preference in blind pairwise user votes.

Top models (higher is better)

ModelScore
Claude Opus 5.51532
Gemini 4 Argon1524
Claude Fable 5.11512
Opus 4.61502
Claude Fable 51501
Claude Opus 51499
Muse Wonder1497
Gemini 3.8 Flash1494
Opus 4.71492
Gemini 3.7 Flash1491
Gemini 3.1 Pro Preview1481
Gemini 3 Pro Preview1478
Gemini 3.5 Flash1477
Qwen3.8-Max1477
Claude Sonnet 5.51476
Muse Spark 1.31470
Kimi K31470
Gemini 3.6 Flash1470
MiMo-V2.6-Pro1470
GLM-5.31469
GPT-5.51466
GPT-5.6 Sol1463
GLM-5.21463
Qwen3.7 Max Preview1462
Opus 4.81461
Gemini 3 Flash Preview1459
GPT-5.41458
GLM-5.3 Flash1455
GLM-5.11455
Muse Spark1454
Gemini 2.5 Pro1454
Muse Spark 1.21454
Opus 4.51452
MiMo-V2.5-Pro1451
DeepSeek V4.1 Flash1451
ERNIE 5.11449
Muse Spark 1.11447
DeepSeek-V4-Pro1447
Grok 4.51445
Sonnet 4.61444
Loading Atlas data…