Browser Use

By Browser Use

Let an LLM agent drive a real browser — navigate, click, type, and extract from any website.

Website ↗ Repository ↗ Docs ↗

v0.13.62026-07-17PythonMIT

#browser #automation #agent #python

106.2k Open source Free tier Paid

Doc version 1.0 · Updated June 28, 2026

Related concepts

Tools Doc version 1.0 · Updated June 29, 2026 The functions that connect a model to the outside world — web search, scraping, browser control, and app integration that bridge to real data and actions beyond the training cutoff.

Related tools

Stagehand

Browser & Computer Use

E2B

Sandboxes & Runtimes

Overview

Browser Use lets an LLM agent drive a real browser to finish web tasks end to end.
You describe the goal in plain English, and the agent navigates, clicks, types, and extracts data from live pages.

flowchart LR
  task[task] --> agent[Browser Use Agent]
  agent --> browser[(real browser)]
  agent --> out[result]

The Code samples tab shows running a task with an LLM-driven agent.

When to use it

Choose Browser Use when an agent has to interact with a site that offers no clean API. It shines for scraping behind logins, filling forms, and multi-step flows a human would click through.

Pricing

Open source and free to self-host. Browser Use Cloud runs the agent on hosted, stealth browsers billed in credits.

Plan	Price	Notes
Free	$0	3 concurrent sessions, community support
Dev	$29/mo	$29 credits, 25 sessions, scheduled jobs
Business	$299/mo	$299 credits, 200 sessions
Scaleup	$999/mo	$999 credits, 500 sessions
Enterprise	Contact	Custom SLAs and limits

Source ↗ · as of 2026-06-27

Example

Python · run a task

When to use

Reach for Browser Use when an agent must operate a real site that has no API.

How it flows

You give a plain-English task; the LLM plans clicks and reads results back.

Structure

flowchart LR
  task[task] --> agent[Browser Use Agent]
  agent --> browser[(real browser)]
  agent --> out[result]

from browser_use import Agent, ChatAnthropic
import asyncio

# The agent pairs with an LLM that decides each browser action.
async def main():
    agent = Agent(
        task="Find the number of stars on the browser-use repo",
        llm=ChatAnthropic(model="claude-opus-4-8"),
    )
    history = await agent.run()
    print(history.final_result())

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