Atlas

Benchmarks

← All benchmarks

LMArena Text Arena Japanese Elo

Chat & Writing · 2024-06-03

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 5.11551
Claude Opus 5.51532
Claude Fable 51521
Gemini 3.7 Flash1513
Kimi K31511
Claude Opus 51509
Gemini 3.8 Flash1508
GPT-5.51505
Gemini 3 Pro Preview1504
Opus 4.61499
GPT-5.6 Sol1497
Gemini 3.1 Pro Preview1496
Opus 4.71490
Gemini 3 Flash Preview1489
GPT-5.41480
Gemini 3.5 Flash1479
Qwen3.8-Max1478
Gemini 3.6 Flash1476
Muse Spark 1.31476
GPT-5.6 Terra1471
Claude Sonnet 5.51468
Muse Spark 1.11468
GLM-5.31467
GPT-5.11461
Opus 4.81461
DeepSeek-V4-Pro1456
GLM-5.21453
DeepSeek V4.1 Flash1453
GLM-5.3 Flash1450
Gemini 3.5 Flash-Lite1445
Gemini 2.5 Pro1444
Grok 4.51444
GPT-6 Sol1443
Sonnet 51443
Kimi K2.61442
Sonnet 4.61441
Opus 4.51441
GLM-4.71440
GPT-6 Astra1437
GPT-5.21436
Loading Atlas data…