Atlas

Benchmarks

← All benchmarks

LMArena Text Arena Occupational Life, Physical, and Social Science Elo

Professional Work · 2025-11-05

Category-specific Elo score from the LMArena text-arena voting leaderboard. Higher Elo indicates stronger crowd-voted preference within this category.

Top models (higher is better)

ModelScore
Gemini 4 Argon1532
Opus 4.61527
Claude Fable 51525
Opus 4.71525
Kimi K31524
Claude Fable 5.11522
Muse Wonder1521
Claude Opus 5.51520
Gemini 3.8 Flash1519
Qwen3.8-Max1514
Claude Opus 51514
Gemini 3.7 Flash1514
Muse Spark 1.31512
Gemini 3.1 Pro Preview1510
Muse Spark 1.21509
GLM-5.31506
Opus 4.81504
Muse Spark 1.11501
Sonnet 4.61500
Claude Sonnet 5.51498
GLM-5.21498
MiMo-V2.6-Pro1498
GPT-5.51498
Gemini 3 Pro Preview1498
Gemini 3.5 Flash1496
Muse Spark1496
Gemini 3.6 Flash1494
GPT-6 Astra1493
GPT-5.6 Sol1490
MiMo-V2.5-Pro1490
DeepSeek V4.1 Flash1490
GLM-5.3 Flash1490
Sonnet 51488
GPT-5.41488
Opus 4.51487
GLM-5.11487
Gemini 3 Flash Preview1487
GPT-6.1 Sol1486
DeepSeek V4 Pro 08131486
ERNIE 5.11486
Loading Atlas data…