Atlas

Benchmarks

← All benchmarks

LMArena Text Arena Occupational Business, Management & Financial Ops 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
Opus 4.71508
Claude Fable 51507
Opus 4.61503
Claude Opus 51496
Kimi K31495
Muse Spark 1.11494
Opus 4.81493
Muse Spark1490
GPT-5.51489
Gemini 3.6 Flash1488
GPT-5.41484
GPT-5.6 Sol1483
Sonnet 4.61482
Gemini 3.1 Pro Preview1478
ERNIE 5.11476
Gemini 3 Pro Preview1476
Opus 4.51472
MiMo-V2.5-Pro1471
Gemini 3 Flash Preview1470
GPT-5.6 Terra1470
Sonnet 51466
GPT-5.6 Luna1466
Grok 4.51465
Grok 4.11462
GLM-5.11462
Gemini 3.5 Flash1462
DeepSeek-V4-Pro1461
Gemini 3.5 Flash-Lite1461
Sonnet 4.51461
GPT-5.4 Mini1460
GPT-5.3 Instant1460
Qwen3.7 Max Preview1458
GLM-5.21456
MiMo-V2-Pro1456
Kimi K2.61456
MiniMax M31453
GLM-51451
Gemma 4 26B A4B1450
Hy31450
Inkling1448
Loading Atlas data…