Browser Use

Open

A web service runs AI agents that carry out browser tasks from text instructions, navigate pages, interact with forms, and return the results.

Description

Browser Use takes a plain-language task and hands it to an AI agent operating a browser. The agent visits websites, interacts with page elements, and reports the outcome.

Developers can use the hosted API to start tasks and receive text or structured responses. An open-source library is also available for running browser agents in their own environments.

It is intended for web automation and retrieving information that requires multiple browser actions. Whether a particular task succeeds depends on website access and the site's terms.

Cloud browser time is billed separately from model tokens used by the agent, which include a service markup. Traffic and proxies can incur additional charges; an ongoing free cloud plan has not been confirmed.