Atlas

Benchmarks

← All benchmarks

LMArena Text Arena Occupational Mathematical Elo

Math · 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 Argon1540
GPT-6.1 Sol1538
Claude Fable 5.11535
Claude Opus 51532
Opus 4.61524
Muse Wonder1519
Claude Fable 51519
Claude Opus 5.51517
Opus 4.71515
Gemini 3.8 Flash1514
Claude Sonnet 5.51513
GPT-5.6 Sol1512
Kimi K31508
Opus 4.81506
GPT-5.51506
GPT-6 Astra1503
MiMo-V2.6-Pro1502
Muse Spark 1.31500
GPT-5.41499
Muse Spark 1.11498
Gemini 3.7 Flash1498
GLM-5.31497
GLM-5.3 Flash1496
DeepSeek V4.1 Flash1496
GLM-5.21494
Grok 4.51494
Hy31493
Qwen3.8-Max1491
GPT-5.6 Terra1491
Muse Spark 1.21491
MiMo-V2.5-Pro1490
Gemini 3.6 Flash1489
ERNIE 5.11489
Gemini 3.5 Flash1489
Sonnet 51489
Grok 4.61486
Gemini 3.1 Pro Preview1485
GPT-5.6 Luna1485
Kimi K2.61484
Sonnet 4.61483
Loading Atlas data…