The Browser Use blog
RSSWhat we learned building this.
Engineering deep dives, experiments, and lessons from building agents for the web.
Explore the blog
39 articles found.
What is a Headless Browser?
Learn what a headless browser is, how it compares to a headful one, its use cases, and the basics of browser automation.
Nik Tomazic
Browser Use can now pay with your credit card using Stripe Link CLI
You can safely connect your credit card and purchase things with hosted browser agents.
Gregor Zunic
How we Made Scalable Long-Horizon RL Environments for Browser Use
Browser agent benchmarks are saturating. This is how we build RL environments from real user traffic against the real web.
Alexander Yue
Browser Agents Can Now Pay With Crypto: Coinbase x Browser Use
Browser Use Cloud now accepts payment in USDC over the x402 protocol. No signup, no credit card, and no API key are needed. Your crypto wallet is your account.
Laith Weinberger
The Benchmark Behind the Benchmark
A benchmark is a dataset and a judge. The judge sounds like a detail. It is the hard part.
Gregor Zunic
How We Built Secure, Scalable Agent Infrastructure with AWS Bedrock AgentCore
Agents should have nothing worth stealing and nothing worth preserving. Here's how we achieve that at scale using AWS Bedrock AgentCore.
Larsen Cundric
How to Use AI Agents for QA Automation
Automate QA with an AI agent: point the QA skill at any site to test it like a user, score it 1–5, and get what to fix. Works in Claude Code, Codex, and Cursor.
Shawn Pana & Laith Weinberger
How We Made Cloud Browsers 3x Cheaper and Faster
We cut browser sessions from $0.06 to $0.02 per hour while making browsers start and scale faster.
Aitor Mato & Laith Weinberger
Claude Fable Sets High Score on BU Bench
We ran Claude Fable 5 on BU Bench with the open-source Browser Use library. Scoring 80%, it bested all other models.
Laith Weinberger
A Python Program To Scrape Any Website
Use our Python SDK to scrape any website instantly and without being blocked.
Laith Weinberger
A Production Architecture for the Browser Use Open-Source Library
The SQS-to-Lambda architecture we run on top of the open-source browser-use library.
Larsen Cundric
Everything I Got Wrong in the Last 4,000 Commits
A year of mistakes as the first engineer at Browser Use, in chronological order.
Larsen Cundric
BUX: Your 24/7 Remote Agent with Browser Harness
Browser Use Box (BUX) is a 24/7 remote VM with Claude Code and Browser Harness pre-installed, controlled from Telegram, the web, or SSH.
Johannes Dittrich
The Bitter Lesson of Agent Harnesses
Don't wrap the LLM. Don't wrap its tools either.
Gregor Zunic
Prove You Are a Robot: The Reverse-CAPTCHA Experiment
How we built a reverse-CAPTCHA for agents, and what we learned.
Luka Secilmis
Remote Browsers for Agents: Browser Use Free Tier
How to give AI agents access to fast, state-of-the-art remote browsers.
Reagan Hsu
Web Agents That Actually Learn
Every agent that visits a website starts from scratch. We made them remember.
Gregor Zunic
The Ultimate Guide to Web Scraping (2026)
I tested the five most popular web scraping tools — Firecrawl, Bright Data, Cloudflare, Browserbase, and Browser Use — to help you pick the right one.
Reagan Hsu
How we built the best browser agent with Auto-Research
97% on Online-Mind2Web benchmark. The highest score ever.
Magnus Müller
We Stealth Benchmarked Every Major Cloud Browser Provider
We built a stealth benchmark from real production data and tested every major cloud browser.
Aitor Mato
How New Generation uses Browser Use to Make the Web Ready for Agents
How New Generation uses Browser Use to score e-commerce sites for agent readiness.
Luka Secilmis
What Happens When You Give your Agent Maximum Freedom
Training beats prompting. Every time.
Gregor Zunic
How to Authenticate AI Web Agents
A comprehensive guide to logging AI web agents into your accounts — from syncing browser profiles to handling 2FA codes.
Laith Weinberger
How to Win Hackathons: A Guide
Winning hackathons is generally difficult, but there are some simple things you can do to place yourself above the rest.
Reagan Hsu
How We Built Secure, Scalable Agent Sandbox Infrastructure
From AWS Lambda to Unikraft micro-VMs with a control plane architecture.
Larsen Cundric
How we built scalable evaluation infrastructure for AI web agents
A deep dive into our highly scalable LLM-as-a-judge evaluation platform that solves web agent variance.
Alexander Yue
What LLM model should I use for Browser Use? The Definitive Browser AI Benchmark
A statistically rigorous benchmark comparing the speed, cost, and accuracy of top frontier models for browser automation.
Alexander Yue
How Browser Use Empowers Frigade to Build the Best Onboarding Agent
How Frigade uses Browser Use to build AI-powered onboarding that understands any product.
Reagan Hsu
Browser agent bot detection is about to change
How we built Browser Use to stay undetectable
Aitor Mato
Browser Agent Benchmark: Comparing LLM Models for Web Automation
We benchmark every major LLM on 100 hard browser tasks. Browser Use Cloud scores 78%, 16 points ahead of the best open-source model.
Alexander Yue
Browser Use <> Parallel AI - The Future of Web Search
How Browser Use and Parallel AI are building the future of web search together.
Reagan Hsu
The Bitter Lesson of Agent Frameworks
All the value is in the RL'd model, not your 10,000 lines of abstractions.
Gregor Zunic
Celebrating One Year of Progress in Browser Agents
One year after our first commit, we reflect on how far browser agents have come—and where they need to go next.
Gregor Zunic
Speed Matters: How Browser Use Achieves the Fastest Agent Execution
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.
Gregor Zunic
The Fastest Web Agent in the World
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?
Gregor Zunic
Stealth Browser Infrastructure
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 👀
Gregor Zunic
Closer to the Metal: Leaving Playwright for CDP
Why we dropped Playwright and switched to raw CDP for faster, more capable browser automation.
We Raised $17M to Build the Future of Web for Agents
How a weekend project turned into the largest open-source browser automation framework—and why we're just getting started.
Gregor Zunic
Browser Use = state of the art Web Agent
Browser Use achieves state-of-the-art performance on the WebVoyager benchmark with an 89.1% success rate.
Gregor Zunic