Atlas

Benchmarks

← All benchmarks

LMArena Text Arena Hard Prompts 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 Argon1540
Claude Opus 5.51535
Opus 4.61530
Claude Fable 51529
Opus 4.71525
Claude Fable 5.11523
Muse Wonder1522
Gemini 3.8 Flash1517
Muse Spark 1.31517
Kimi K31516
Claude Opus 51516
Opus 4.81513
Muse Spark 1.21512
GPT-5.6 Sol1511
Muse Spark 1.11510
Gemini 3.7 Flash1509
Gemini 3.1 Pro Preview1508
Muse Spark1506
GLM-5.31505
Claude Sonnet 5.51504
GPT-6.1 Sol1504
Qwen3.8-Max1504
MiMo-V2.6-Pro1504
Sonnet 4.61503
Gemini 3 Pro Preview1502
GPT-5.51501
Gemini 3.6 Flash1501
GPT-6 Astra1501
GLM-5.3 Flash1501
DeepSeek V4.1 Flash1499
Opus 4.51498
GPT-5.41497
GLM-5.21496
Qwen3.7 Max Preview1495
Gemini 3.5 Flash1495
MiMo-V2.5-Pro1494
Gemini 3 Flash Preview1493
GPT-5.6 Terra1490
Sonnet 51490
GLM-5.11489
Loading Atlas data…