TencentQQGYLab/AppAgent
A multimodal agent framework enabling LLMs to autonomously navigate and interact with smartphone applications.
Not currently ranked — collecting fresh signals.
star history
AppAgent is a multimodal agent system designed to operate smartphone applications using large language models. It provides a framework where agents can perceive the smartphone environment, make decisions, and execute actions to complete tasks on mobile apps. The system uses vision-language models to understand GUI elements and navigates through apps autonomously without requiring custom API integrations.
Frequently asked
- What is TencentQQGYLab/AppAgent?
- A multimodal agent framework enabling LLMs to autonomously navigate and interact with smartphone applications.
- Is AppAgent open source?
- Yes — TencentQQGYLab/AppAgent is open source, released under the MIT license.
- What language is AppAgent written in?
- TencentQQGYLab/AppAgent is primarily written in Python.
- How popular is AppAgent?
- TencentQQGYLab/AppAgent has 6.8k stars on GitHub.
- Where can I find AppAgent?
- TencentQQGYLab/AppAgent is on GitHub at https://github.com/TencentQQGYLab/AppAgent.