Atlas

Benchmarks

← All benchmarks

LMArena Text Arena Occupational Software & IT Services 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 Argon1549
Opus 4.71540
Claude Fable 51538
Claude Opus 5.51538
Opus 4.61537
Muse Wonder1534
Muse Spark 1.21532
Claude Fable 5.11530
GPT-6.1 Sol1528
MiMo-V2.6-Pro1528
Muse Spark 1.31528
Muse Spark 1.11525
Claude Opus 51523
Opus 4.81523
Kimi K31523
GPT-5.6 Sol1523
GPT-6 Astra1523
Muse Spark1523
Gemini 3.8 Flash1518
Claude Sonnet 5.51516
GLM-5.3 Flash1515
Sonnet 4.61514
DeepSeek V4.1 Flash1514
GLM-5.31513
Gemini 3.1 Pro Preview1513
Qwen3.8-Max1512
Gemini 3.7 Flash1511
GPT-5.51511
Gemini 3.6 Flash1510
Qwen3.7 Max Preview1510
Gemini 3 Pro Preview1509
MiMo-V2.5-Pro1509
ERNIE 5.11507
Opus 4.51507
GPT-5.6 Terra1507
GPT-5.41507
GLM-5.21506
GPT-6 Sol1505
Sonnet 51505
MiMo-V2.6-Flash1503
Loading Atlas data…