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 3.5 Flash1538
Claude Opus 51511
Gemini 3 Pro Preview1502
Gemini 3.1 Pro Preview1500
Gemini 3 Flash Preview1496
Opus 4.61494
Claude Fable 51490
Opus 4.71486
Muse Spark 1.11483
Gemini 2.5 Pro1480
Muse Spark1474
GPT-5.41472
MiMo-V2.5-Pro1467
Opus 4.81467
GPT-5.51466
ERNIE 5.11465
GLM-5.11461
DeepSeek-V4-Pro1459
Grok 4.11457
Qwen3.7-Plus1454
Kimi K2.51446
Qwen3.5 397B A17B1445
GLM 4.61445
GPT-5.11444
Gemini 3.1 Flash-Lite Preview1443
Sonnet 4.61443
Qwen3 Max Preview1441
Kimi K2.61440
O3 (2025-04-16)1440
GLM-5.21438
GLM-51438
GPT-5.6 Sol1436
MiniMax M31436
Opus 4.51434
Qwen3.6 Plus (2026-04-02)1433
Mistral Medium 3.11433
DeepSeek-V4-Flash1432
DeepSeek-R1-05281431
GLM 5V Turbo1431
GPT-5.21430
Loading Atlas data…