Atlas

Benchmarks

← All benchmarks

Arena Text — German (No Style Control)

Chat & Writing

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

Top models (higher is better)

ModelScore
Claude Opus 51530
Gemini 3 Pro Preview1515
Muse Spark 1.31513
Gemini 3.7 Flash1509
Gemini 3.8 Flash1509
Opus 4.61502
Claude Fable 5.11498
Muse Spark1498
Gemini 3 Flash Preview1497
Opus 4.71495
Gemini 3.5 Flash1494
Kimi K31491
Gemini 3.1 Pro Preview1490
Gemini 2.5 Pro1487
GLM-5.31486
Claude Fable 51485
Gemini 3.6 Flash1481
GPT-5.51480
Qwen3.8-Max1476
GPT-5.6 Sol1475
Nemotron 3 Ultra 550B A55B (NVFP4)1474
Opus 4.81473
GPT-5.41473
GLM-5.21471
Qwen3.7-Plus (source unspecified)1471
ERNIE 5.11470
Muse Spark 1.11469
DeepSeek V4.1 Flash1468
GLM-5.11467
Qwen3 Max Preview1462
GPT-5.6 Terra1461
MiMo-V2.5-Pro1459
DeepSeek-V4-Pro1458
GLM-5.3 Flash1457
GPT-5.6 Luna1455
MiMo-V2.6-Flash1455
GPT-6 Astra1455
Gemini 3.5 Flash-Lite1454
Qwen3.6 Plus (2026-04-02)1452
Step 5 Preview1451
Loading Atlas data…