All the value is in the RL'd model, not your 10,000 lines of abstractions.
One year after our first commit, we reflect on how far browser agents have come—and where they need to go next.
Matching state-of-the-art accuracy is table stakes. We built BU 1.0 to be the fastest browser agent while maintaining the same performance.
We built a special LLM gateway that reduces the latency by 6x while keeping the same performance. The agents can now take 20 steps per minute. How many can a human do?
Introducing: Stealth and native browser use infrastructure built directly into the library. Want to bypass Cloudflare, or any other anti-bot protection? It's never been easier 👀
Our journey migrating off Playwright in persuit of finer-grained control of the browser.
How a weekend project turned into the largest open-source browser automation framework—and why we're just getting started.
Browser Use has achieved state-of-the-art performance on the WebVoyager benchmark, with an impressive **89.1% success rate** across 586 diverse web tasks.