Atlas

Benchmarks

← All benchmarks

LMArena Text Arena Creative Writing Elo

Chat & Writing · 2024-10-30

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
Claude Opus 5.51520
Gemini 4 Argon1512
Claude Fable 51501
Opus 4.61499
Muse Wonder1492
Gemini 3.7 Flash1492
Opus 4.71489
Claude Fable 5.11487
Gemini 3.8 Flash1485
Gemini 3 Pro Preview1485
Gemini 3.1 Pro Preview1481
Claude Opus 51474
Gemini 3.6 Flash1471
GPT-5.6 Sol1470
Opus 4.81469
Gemini 3.5 Flash1469
Qwen3.8-Max1468
Claude Sonnet 5.51466
Muse Spark1464
Muse Spark 1.31463
Kimi K31462
Opus 4.51462
Gemini 3 Flash Preview1457
GLM-5.31456
Sonnet 4.51454
GPT-6.1 Sol1453
GLM-5.21453
Grok 4.51452
GPT-5.51452
GPT-6 Astra1449
Sonnet 4.61449
GLM-51448
DeepSeek-V4-Pro1446
GLM-5.11446
MiMo-V2.6-Pro1446
DeepSeek V4 Pro 08131446
Gemini 2.5 Pro1443
Grok 4.61443
Muse Spark 1.11443
GPT-5.41442
Loading Atlas data…