Atlas

Benchmarks

← All benchmarks

Arena Text — Polish (Style Controlled)

Chat & Writing

Style-controlled Arena rating for Polish-language Text Arena prompts. Higher ratings indicate stronger preference in blind pairwise user votes after adjusting for response style.

Top models (higher is better)

ModelScore
Claude Fable 5.11518
Gemini 4 Argon1516
Claude Fable 51510
Muse Spark 1.11509
Gemini 3.5 Flash1506
Gemini 3 Pro Preview1505
Claude Opus 51504
Opus 4.61502
Opus 4.71501
Gemini 3.7 Flash1499
Gemini 3 Flash Preview1497
GPT-5.6 Sol1496
Gemini 3.1 Pro Preview1495
Gemini 3.8 Flash1493
Grok 4.11486
Opus 4.81485
Muse Spark 1.31485
Muse Spark1484
GPT-5.41482
GPT-5.51480
GLM-5.31476
Kimi K31475
GPT-5.3 Instant (source unspecified)1475
Qwen3.8-Max1473
Gemini 3.6 Flash1473
Grok 4.71473
DeepSeek V4.1 Flash1472
Sonnet 4.61471
Opus 4.51470
MiMo-V2.5-Pro1467
DeepSeek-V4-Pro1465
GPT-6 Astra1465
GPT-5.6 Terra1464
GPT-5.11463
ERNIE 5.11463
Hy31463
GPT-5.4 Mini1462
Grok 4.61462
GLM-5.21460
Gemini 3.5 Flash-Lite1459
Loading Atlas data…