Atlas

Benchmarks

← All benchmarks

Arena Text — Polish (No Style Control)

Chat & Writing

Raw Arena rating for Polish-language Text Arena prompts without style-control adjustment. Higher ratings indicate stronger preference in blind pairwise user votes.

Top models (higher is better)

ModelScore
Gemini 4 Argon1530
Claude Fable 5.11530
Claude Opus 51523
Gemini 3.5 Flash1517
Gemini 3 Pro Preview1503
Gemini 3.7 Flash1503
Opus 4.61497
Muse Spark 1.11497
Gemini 3.1 Pro Preview1497
Gemini 3.8 Flash1496
Gemini 3 Flash Preview1494
Claude Fable 51490
Opus 4.71489
Muse Spark 1.31487
Gemini 2.5 Pro1479
Gemini 3.6 Flash1475
Muse Spark1473
GPT-5.41472
Qwen3.8-Max1471
ERNIE 5.11467
GLM-5.31466
GPT-5.51466
MiMo-V2.5-Pro1463
DeepSeek-V4-Pro1460
Kimi K31460
Opus 4.81459
GPT-5.6 Sol1458
GLM-5.21457
DeepSeek V4.1 Flash1457
Grok 4.11454
GLM-5.11452
Inkling1450
GLM-5.3 Flash1450
Qwen3.7-Plus (source unspecified)1449
Qwen3.5 397B A17B1448
GPT-5.6 Terra1447
Qwen3.8 27B1447
GLM 4.61447
MiMo-V2.6-Flash1446
Hy31446
Loading Atlas data…