Browser Use Web UI is a Gradio-based interface for running a browser-use agent from a web page. It exposes most of the underlying browser-use functionality, so you can type a task, pick a model, and watch the agent operate a browser without writing code.
You run it locally or through Docker Compose, and the Docker setup adds a VNC viewer for watching the browser. The agent can drive a fresh Playwright-installed browser or your own Chrome profile, which keeps existing logins and avoids re-authentication, and can record the screen in high definition.
Features
- Gradio interface: configure and launch browser-use agents from a web form
- Model providers: Google, OpenAI, Azure OpenAI, Anthropic, DeepSeek, Ollama, and others, including DeepSeek-R1 for reasoning-heavy tasks
- Own-browser mode: run the agent in your existing Chrome profile
- Persistent sessions: keep the browser window open between tasks to preserve state and history
- Screen recording: high-definition recording of agent sessions
- Docker deployment: Compose setup with a noVNC viewer, including an ARM64 build path