Skip to main content
Browser Use Cloud runs AI browser agents on stealth cloud browsers. You describe a task in plain language and the agent drives the browser to completion. Because the agent runs on a real cloud browser, you can also connect to that same browser yourself over CDP whenever you want hands-on control — the agent and the browser are one system, not a choice between two. Every session — whether the agent is driving or you are — runs a hardened Chromium fork with anti-fingerprinting, automatic CAPTCHA solving, and residential proxies enabled by default.

​
Run a task

Describe the outcome and let the agent handle the clicks — price monitoring, form filling, research, multi-step workflows:

​
Drive the browser yourself

The agent runs on a real cloud browser, and you can connect to it over CDP with Playwright, Puppeteer, or any framework — for exact selectors and timing, or to build your own agent on top:
Mix the two in one session: let the agent handle the fuzzy steps and your code handle the deterministic ones.

​
Already using the open-source library?

API v2 is the open-source library running on our infrastructure instead of your own machine — the same agent code, now on managed stealth browsers with no local setup. It also gives you our bu-1-0 and bu-2-0 models, which are cheaper, faster, and more accurate than general-purpose models on browser tasks.

​
Start here

​
Further reading