X-PLUG/MobileAgent
A multimodal AI agent framework that automates mobile GUI interactions using vision-language models.

Velocity · 7d
+10
★ / day
Trend
→steady
star history
Mobile-Agent is an autonomous agent system designed to control mobile device interfaces through multimodal large language models. It leverages visual understanding to navigate apps, execute tasks, and perform GUI automation without direct user input. The project includes specialized GUI-Owl models (7B and 32B variants) trained for mobile interface understanding and control.