Atlas

Benchmarks

← All benchmarks

LMArena Text Arena Korean Elo

Chat & Writing · 2024-06-29

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
Claude Fable 5.11499
Claude Opus 51496
Claude Fable 51496
Muse Spark 1.31493
Gemini 3.7 Flash1481
Muse Spark 1.11474
Muse Spark1474
Kimi K31472
Opus 4.71470
Muse Spark 1.21465
DeepSeek V4 Pro 08131464
GLM-5.31463
Opus 4.61463
Gemini 3.8 Flash1460
Qwen3.8-Max1456
GPT-5.6 Sol1455
Gemini 3.1 Pro Preview1455
GPT-6 Astra1451
Gemini 3 Pro Preview1449
Opus 4.81448
Opus 4.51447
DeepSeek V4.1 Flash1446
Gemini 3 Flash Preview1445
GPT-5.51444
GLM-5.21440
Gemini 3.5 Flash1439
GLM-5.3 Flash1438
MiMo-V2.5-Pro1436
Sonnet 51433
GPT-5.41433
Sonnet 4.61432
Grok 4.11431
Gemini 3.6 Flash1429
GLM-51426
Grok 4.71425
Gemini 3.5 Flash-Lite1424
Kimi K2.61423
MiMo-V2.6-Flash1423
GPT-5.6 Terra1422
ERNIE 5.11421
Loading Atlas data…