Atlas

Benchmarks

← All benchmarks

Arena Text — Exclude Ties (No Style Control)

Chat & Writing

Raw Text Arena rating computed with tied votes excluded and without style-control adjustment. It is separate because both filters change the vote sample and rating estimates.

Top models (higher is better)

ModelScore
Gemini 4 Argon1551
Claude Opus 5.51539
Claude Fable 5.11534
Claude Opus 51527
Muse Wonder1519
Opus 4.61517
Gemini 3.8 Flash1514
Muse Spark 1.31505
Claude Fable 51503
MiMo-V2.6-Pro1503
Gemini 3.7 Flash1502
Opus 4.71502
Qwen3.8-Max1494
Gemini 3 Pro Preview1491
Muse Spark 1.21491
Gemini 3.5 Flash1491
Gemini 3.1 Pro Preview1490
Gemini 3.6 Flash1488
Muse Spark 1.11486
Kimi K31483
GLM-5.31480
Qwen3.7 Max Preview1480
Muse Spark1480
Claude Sonnet 5.51479
GLM-5.3 Flash1478
GPT-5.51477
GLM-5.21475
GPT-5.41474
Gemini 3 Flash Preview1474
ERNIE 5.11472
DeepSeek V4.1 Flash1468
MiMo-V2.5-Pro1466
GLM-5.11465
Opus 4.81464
Gemini 2.5 Pro1460
GPT-5.6 Sol1459
Sonnet 4.61458
Kimi K2.61455
Qwen3.7-Plus (source unspecified)1454
MiMo-V2.6-Flash1450
Loading Atlas data…