Atlas

Benchmarks

← All benchmarks

LMArena Text Math No Style Control Elo

Math · 2024-09-09

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 Argon1538
Claude Opus 51533
Claude Fable 5.11532
Gemini 3.8 Flash1532
Claude Opus 5.51518
Opus 4.61518
Claude Fable 51516
Gemini 3.7 Flash1510
Gemini 3.6 Flash1507
Gemini 3.5 Flash1504
Muse Spark 1.31502
GLM-5.3 Flash1502
Qwen3.8-Max1501
MiMo-V2.6-Pro1500
Opus 4.71498
Claude Sonnet 5.51498
GLM-5.31498
Kimi K31493
GPT-6.1 Sol1490
Qwen3.7 Max Preview1488
GPT-5.41488
GPT-5.51487
Opus 4.81486
Gemini 3.1 Pro Preview1486
GLM-5.21484
Muse Wonder1484
Muse Spark 1.21483
ERNIE 5.11481
Muse Spark 1.11481
Inkling1481
MiMo-V2.5-Pro1480
DeepSeek V4.1 Flash1479
Gemini 3 Pro Preview1476
Hy31476
Kimi K2.61474
GLM-5.11474
GPT-5.6 Sol1473
Gemini 3 Flash Preview1473
Kimi K2.51470
Gemma 4 26B A4B IT1470
Loading Atlas data…