Browser Use

Agent-nativeFreemium

Stealth browser automation with persistent sessions for agents

Visit website GitHub

Overview

About Browser Use

This summary uses submitted and public product material. Only source-mapped claims are documentation reviewed.

Browser Use combines an open-source browser agent library with managed cloud infrastructure. Its current cloud offering emphasizes stealth browsers, persistent sessions and profiles, residential proxies, CAPTCHA solving, and both natural-language tasks and raw browser sessions.

Features

Natural-language browser tasks that return structured data
Stealth browsers with anti-detect fingerprinting and automatic CAPTCHA solving
Residential proxies in 195+ countries with session-level location control
Persistent sessions and profiles for login state and multi-step workflows
Raw browser sessions over CDP for lower-level automation when needed

So agents can…

Fill forms, log in, and complete multi-page workflows on JavaScript-heavy sites
Keep authenticated browser state across runs instead of starting fresh each time
Handle protected sites that would otherwise block standard headless automation

Listing decision

Why Browser Use is listed

Browser Use supplies the model-driven browser agent loop itself, accepting goals and selecting browser actions through an LLM rather than only exposing a neutral browser-control primitive.

Before you choose

Fit and limitations

Best for

Developers building agents that must navigate dynamic sites, complete multi-page workflows, reuse authenticated state, or scale browser sessions through either an open-source local runtime or managed cloud.

Known limitations

Local production use must account for Chrome memory and concurrency. Cloud sessions have a four-hour timeout, and profile changes persist only when the session is explicitly stopped. CAPTCHA handling and stronger fingerprinting are positioned as cloud capabilities rather than guarantees of the local library.

Not yet verified

First-party speed, accuracy, stealth, proxy-coverage, and CAPTCHA claims were not independently benchmarked or hands-on tested for this record; success remains site- and model-dependent.

Sources

Sources and verification

First-party unless marked otherwise. Each link supports only its adjacent claim.

Browser Use open-source documentation — local and self-hosted agent library
Browser Use documents a Python AI browser-automation library that can connect an LLM and run locally or self-hosted.
official documentation · accessed Sep 3, 2026
Browser Use repository — SDK, CLI, cloud, and production caveats
The maintained repository documents the Python agent and CLI, optional cloud service, supported local use, and resource-management reasons for using managed infrastructure at scale.
official repository · accessed Sep 3, 2026
Browser Use pricing — cloud rates, concurrency, and deployment tiers
The cloud pricing page publishes pay-as-you-go and subscription tiers, browser and proxy rates, concurrency limits, four-hour session timeouts, and custom on-premises deployment.
official pricing · accessed Sep 3, 2026
Browser Use profiles guide — persistence requirement
Cloud profiles persist cookies and browser storage, but changes are saved only when the session is explicitly stopped rather than left open or allowed to time out.
official documentation · accessed Sep 3, 2026
Browser Use license — MIT
The open-source Browser Use repository is published under the MIT License.
official license · accessed Sep 3, 2026