Atlas

Benchmarks

← All benchmarks

Arena Vision — Creative Writing (No Style Control)

Multimodal

Raw Vision Arena rating for visually grounded creative-writing prompts without style-control adjustment. Higher ratings indicate stronger preference in blind pairwise user votes.

Top models (higher is better)

ModelScore
Gemini 3 Pro Preview1342
Claude Fable 51338
Opus 4.61331
Claude Fable 5.11330
Muse Spark1327
Opus 4.71326
Gemini 3.1 Pro Preview1324
Gemini 3 Flash Preview1314
Claude Opus 51313
Qwen3.8-Max1312
Gemini 3.7 Flash1312
GPT-5.51301
Gemini 3.5 Flash1301
Opus 4.81301
GPT-6.1 Sol1301
Gemini 3.8 Flash1299
Muse Spark 1.31296
Grok 4.51293
GPT-6 Astra1286
Gemma 4 31B IT1285
GPT-5.6 Sol1283
GPT-5.41283
Gemini 3.6 Flash1278
DeepSeek V4.1 Flash1275
Kimi K2.61273
Sonnet 4.61272
Gemini 2.5 Flash Preview (09-2025)1270
GLM-5.3 Flash1270
Kimi K2.51269
GPT-5.11266
Gemini 2.5 Pro1265
Grok 4.61262
Qwen3.5 397B A17B1260
Gemini 3.5 Flash-Lite1260
Qwen3.7-Plus (source unspecified)1258
GPT-5.6 Luna1257
Gemma 4 26B A4B IT1256
Muse Spark 1.11253
GLM 5V Turbo1252
GPT-5.6 Terra1248
Loading Atlas data…