Atlas

Benchmarks

← All benchmarks

LMArena Text Arena Hard Prompts English Elo

Chat & Writing · 2024-05-17

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 Argon1538
Claude Opus 5.51537
Opus 4.61534
Claude Fable 51531
Opus 4.71528
Muse Wonder1523
Claude Fable 5.11521
Kimi K31517
Opus 4.81517
Muse Spark 1.31516
Gemini 3.8 Flash1514
Claude Opus 51513
MiMo-V2.6-Pro1513
Claude Sonnet 5.51513
GLM-5.31511
Qwen3.8-Max1510
GPT-6.1 Sol1510
Muse Spark 1.21509
GPT-5.6 Sol1509
Muse Spark1508
GPT-6 Astra1508
Gemini 3.1 Pro Preview1507
Sonnet 4.61506
Muse Spark 1.11505
DeepSeek V4.1 Flash1504
MiMo-V2.5-Pro1500
GPT-5.51500
Gemini 3.6 Flash1500
Gemini 3 Pro Preview1500
Gemini 3.7 Flash1499
GLM-5.11499
Opus 4.51499
GLM-5.3 Flash1499
GLM-5.21498
Sonnet 51494
GPT-5.41493
Grok 4.51493
ERNIE 5.11491
Gemini 3 Flash Preview1491
Gemini 3.5 Flash1491
Loading Atlas data…