Atlas

Benchmarks

← All benchmarks

LMArena Text Arena Chinese Elo

Chat & Writing · 2024-04-14

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 Argon1579
Claude Opus 5.51570
Claude Fable 5.11568
Claude Fable 51555
Opus 4.61551
Gemini 3.7 Flash1546
Claude Opus 51546
Muse Wonder1544
Claude Sonnet 5.51542
Kimi K31542
GPT-5.6 Sol1541
Gemini 3.8 Flash1538
GLM-5.31538
Qwen3.8-Max1537
Opus 4.71537
GLM-5.3 Flash1532
Muse Spark 1.31531
Gemini 3.6 Flash1530
Muse Spark 1.11529
Gemini 3.1 Pro Preview1528
Opus 4.81525
GPT-5.51525
Muse Spark1525
GLM-5.21523
DeepSeek V4.1 Flash1522
Gemini 3 Pro Preview1521
Qwen3.7 Max Preview1520
Gemini 3.5 Flash1518
GPT-5.6 Terra1517
Gemini 3 Flash Preview1517
Kimi K2.61516
GLM-51516
GPT-6.1 Sol1513
MiMo-V2.5-Pro1513
GLM-5.11513
GPT-6 Astra1512
MiMo-V2.6-Pro1511
Sonnet 4.61509
Muse Spark 1.21509
GPT-5.41506
Loading Atlas data…