Browser Use

Coding Agents 💻 Python ⚖️ MIT 🟢 Actively maintained
112.4k stars

Browser Use is an open-source Python library and CLI skill that enables AI agents to control a web browser like a human—clicking, typing, and filling forms to complete tasks. It solves the problem of automating web interactions without brittle scripts, and is for developers building web automation or integrating browser control into AI agents.

Browser Use: This New AI Agent Can Do Anything (Full AI Scraping Tutorial)

🎬 Browser Use: This New AI Agent Can Do Anything (Full AI Scraping Tutorial) · Tech With Tim

Browser Use demo
🖼️ Screenshot from the project README

✨ Key features

  • AI agent controls browser like a human
  • Fill forms, click buttons, type text
  • Extract structured data from web pages
  • Works with any LLM via provider-prefixed models
  • Open-source with MIT license
  • Cloud-hosted agent with stealth and scaling

🎯 Use cases

  • Automate job applications by filling forms
  • Scrape and export data to CSV
  • Run browser tasks in parallel for monitoring or QA
  • Embed a browser agent into your own product
  • Let an agent upload videos or compare products

📦 Installation

🧰 Requirements: Python >= 3.11, an LLM API key (e.g., BROWSER_USE_API_KEY or provider key), and a browser.

uv add browser-use
# or: pip install browser-use

🚀 Usage

import asyncio
from browser_use import Agent, ChatBrowserUse

async def main():
    agent = Agent(
        task="Find the number of stars of the browser-use repo",
        llm=ChatBrowserUse(model='openai/gpt-5.5'),
    )
    history = await agent.run()

if __name__ == "__main__":
    asyncio.run(main())

⚠️ Good to know

The open-source agent is free but less powerful than the cloud agent for complex tasks; CAPTCHA solving and stealth require the cloud service.

❓ FAQ

Should I use the CLI or the Python library?

Use the CLI if you already have an agent (like Claude Code) that needs to perform browser tasks. Use the Python library when building software that automates the web at scale.

What's the best model to use?

ChatBrowserUse() is optimized for browser automation and completes tasks 3-5x faster than other models with SOTA accuracy. You can also use other providers via model prefixes.

Can I use Claude / GPT / Gemini through ChatBrowserUse?

Yes, ChatBrowserUse accepts provider-prefixed model ids like anthropic/claude-sonnet-4-6 or openai/gpt-5.5, so a single BROWSER_USE_API_KEY reaches all of them.

Can I use this for free?

Yes, Browser Use is open source and free to use. You only need to choose an LLM provider, which may have its own costs.

📊 Repository

Stars★ 112,376
Forks🍴 12,354
Open issues🐛 397
Last commit🕒 Sep 4, 2026
Created📅 Oct 2024
Language💻 Python
License⚖️ MIT

🤖 Overview, features, install steps and FAQ were generated from the project's README on Sep 4, 2026. Always check the original source before running commands.