Atlas

Benchmarks

← All benchmarks

Arena Text — Entertainment, Sports & Media (No Style Control)

Professional Work

Raw Arena rating for entertainment, sports, and media industry prompts without style-control adjustment. Higher ratings indicate stronger preference in blind pairwise user votes.

Top models (higher is better)

ModelScore
Opus 4.61495
Claude Fable 51485
Claude Opus 51480
Opus 4.71473
Gemini 3 Pro Preview1464
Gemini 3.1 Pro Preview1461
Gemini 3.5 Flash1457
Muse Spark1449
Gemini 3 Flash Preview1449
Gemini 3.6 Flash1448
Qwen3.7 Max Preview1448
GPT-5.6 Sol1447
GPT-5.51447
GLM-5.21444
GLM-5.11444
Opus 4.81442
Gemini 2.5 Pro1441
Muse Spark 1.11440
MiMo-V2.5-Pro1439
GPT-5.41439
Kimi K31438
ERNIE 5.11436
Grok 4.51433
Sonnet 4.61432
DeepSeek-V4-Pro1431
Opus 4.51431
Sonnet 4.51427
Qwen3.7-Plus1427
GLM-51426
Kimi K2.61425
Kimi K2.51419
MiMo-V2-Pro1415
GPT-5.11413
GPT-5.6 Terra1412
Gemini 3.5 Flash-Lite1412
GPT-5.6 Luna1409
GLM 4.61409
Grok 4.11409
Sonnet 51408
Gemma 4 26B A4B1407
Loading Atlas data…