Atlas

Benchmarks

← All benchmarks

LMArena Text Arena Math Elo

Math · 2024-07-30

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
Claude Fable 5.11529
Gemini 3.8 Flash1526
Claude Fable 51521
Claude Opus 51521
Opus 4.61517
GPT-6.1 Sol1513
Claude Opus 5.51512
Muse Spark 1.31510
Gemini 3.7 Flash1505
Gemini 3.6 Flash1504
Opus 4.71504
GLM-5.3 Flash1504
Claude Sonnet 5.51501
GPT-5.51501
GLM-5.31500
Kimi K31499
Qwen3.8-Max1499
Gemini 3.5 Flash1497
GPT-5.6 Sol1496
Muse Wonder1496
DeepSeek V4.1 Flash1495
Opus 4.81494
Muse Spark 1.11493
GPT-5.41491
Muse Spark 1.21491
Gemini 3.1 Pro Preview1489
MiMo-V2.6-Pro1488
GLM-5.21488
GPT-6 Astra1487
Qwen3.7 Max Preview1484
Hy31484
GPT-5.6 Terra1483
Inkling1480
MiMo-V2.5-Pro1479
Kimi K2.61479
GPT-5.6 Luna1478
ERNIE 5.11478
Gemini 3 Pro Preview1477
GLM-5.11476
Loading Atlas data…