Atlas

Benchmarks

← All benchmarks

LMArena Text Arena Occupational Legal & Government 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
Kimi K31539
Claude Opus 51519
Opus 4.61511
Opus 4.71510
Claude Fable 51507
Muse Spark1504
Gemini 3 Pro Preview1503
Muse Spark 1.11498
Opus 4.81498
Gemini 3.1 Pro Preview1497
GPT-5.51496
Gemini 3.6 Flash1495
GPT-5.6 Terra1492
GPT-5.41491
GPT-5.6 Sol1490
Gemini 3 Flash Preview1487
Sonnet 4.61484
Gemini 3.5 Flash1484
Opus 4.51483
Sonnet 51480
GLM-5.21480
MiMo-V2.5-Pro1478
Grok 4.11477
GLM-5.11476
Qwen3.7 Max Preview1475
DeepSeek-V4-Pro1472
GPT-5.6 Luna1470
Grok 4.51468
Gemini 2.5 Pro1468
GLM-51467
GPT-5.3 Instant1465
ERNIE 5.11464
Gemma 4 26B A4B1463
Sonnet 4.51463
Kimi K2.61461
DeepSeek-V3.1-Terminus1461
Qwen3.7-Plus1460
GPT-5.11459
Gemma 4 31B1459
MiMo-V2-Pro1458
Loading Atlas data…