Atlas

Benchmarks

← All benchmarks

Arena Vision — OCR (No Style Control)

Multimodal

Raw Vision Arena rating for OCR-centered prompts without style-control adjustment. Higher ratings indicate stronger preference in blind pairwise user votes.

Top models (higher is better)

ModelScore
Claude Fable 51345
Claude Opus 51329
Opus 4.61327
Opus 4.71325
Gemini 3.6 Flash1314
Gemini 3.5 Flash1312
Gemini 3 Pro Preview1311
GPT-5.41310
Muse Spark1308
Gemini 3.1 Pro Preview1306
GPT-5.51306
Opus 4.81305
Grok 4.51301
Muse Spark 1.11300
Sonnet 51300
Sonnet 4.61295
Kimi K2.61291
Gemini 3 Flash Preview1290
GPT-5.6 Sol1286
Qwen3.7-Plus1286
GPT-5.6 Terra1283
Gemini 3.5 Flash-Lite1280
Gemma 4 31B1278
Kimi K2.51276
Qwen3.5 397B A17B1274
Gemini 2.5 Pro1269
GLM 5V Turbo1268
Gemma 4 26B A4B1267
MiniMax M31266
MiMo-V2.51264
GPT-5.4 Mini1257
GPT-5.21256
Gemini 2.5 Flash Preview (09-2025)1256
GPT-5.6 Luna1256
GPT-5.11252
Qwen3.5 122B A10B1252
Gemini 3.1 Flash-Lite Preview1247
MiMo-V2-Omni1239
Gemini 2.5 Flash1238
Grok 4.31237
Loading Atlas data…