List your company

GUI world modeling on MobileWorldBench

2.78AccuracyMobileWorldModel-8B
1.8962.12552.3552.5845May 11, 2026

Evaluation results

MethodLinks
MobileWorldModel-8Bvisual hint=true, VLM-as-a-judge=Gemini-3.1-Pro2026.052.782.892.78.37
MobileWorldModel-8Bvisual hint=false, VLM-as-a-judge=Gemini-3.1-Pro2026.052.662.762.648.07
GPT-5.2VLM-as-a-judge=Gemini-3.1-Pro2026.052.622.483.028.11
Gemini-3.1-ProVLM-as-a-judge=Gemini-3.1-Pro2026.052.572.422.977.96
MobileWorldModel-4Bvisual hint=true, VLM-as-a-judge=Gemini-3.1-Pro2026.052.492.62.57.6
MobileWorldModel-4Bvisual hint=false, VLM-as-a-judge=Gemini-3.1-Pro2026.052.472.592.437.48
Claude-4.6-SonnetVLM-as-a-judge=Gemini-3.1-Pro2026.052.192.082.46.67
Qwen3-VL-4BVLM-as-a-judge=Gemini-3.1-Pro2026.051.981.832.15.83
Qwen3-VL-8BVLM-as-a-judge=Gemini-3.1-Pro2026.051.982.021.975.97
Qwen3-VL-32BVLM-as-a-judge=Gemini-3.1-Pro2026.051.931.922.115.95