zai-org/CogAgent
An open-sourced Vision Language Model-based agent that autonomously perceives and acts on graphical user interfaces to automate computer tasks.
Not currently ranked — collecting fresh signals.
star history
CogAgent is an end-to-end GUI agent built on vision-language models. It perceives screen content, performs reasoning, and executes actions to automate computer tasks across arbitrary graphical interfaces. The project provides model weights, inference code, and training pipelines for developing autonomous agents capable of computer use.
Frequently asked
- What is zai-org/CogAgent?
- An open-sourced Vision Language Model-based agent that autonomously perceives and acts on graphical user interfaces to automate computer tasks.
- Is CogAgent open source?
- Yes — zai-org/CogAgent is open source, released under the Apache-2.0 license.
- What language is CogAgent written in?
- zai-org/CogAgent is primarily written in Python.
- How popular is CogAgent?
- zai-org/CogAgent has 1.2k stars on GitHub.
- Where can I find CogAgent?
- zai-org/CogAgent is on GitHub at https://github.com/zai-org/CogAgent.