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 51520
GPT-5.51513
Gemini 3 Pro Preview1510
Gemini 3.1 Pro Preview1503
Gemini 3 Flash Preview1493
Opus 4.61491
Opus 4.71487
GPT-5.41478
GLM-5.21470
DeepSeek-V4-Pro1467
Sonnet 51462
Gemini 3.5 Flash1462
GPT-5.11462
Gemini 2.5 Pro1452
Kimi K2.61448
Opus 4.81444
Sonnet 4.61444
Opus 4.51441
ERNIE 5.11434
GPT-5.21432
Qwen3.5 397B A17B1432
Grok 4.11431
GPT-51429
O3 (2025-04-16)1427
Kimi K2.51426
MiMo-V2.5-Pro1426
GLM-5.11423
Gemini 3.1 Flash-Lite Preview1422
Grok 4.31419
GPT-5.3 Instant1417
GPT-5.4 Mini1413
MiniMax M31410
Sonnet 4.51410
Qwen3.7-Plus1409
GLM-51408
MiMo-V2-Pro1407
Opus 4.11405
Gemini 2.5 Flash Preview (09-2025)1400
DeepSeek-V4-Flash1396
GLM-4.51395
Loading Atlas data…