Deploy Browser Use

Deploy and Host Browser Use on Railway

Browser Use is an AI-powered browser automation runtime that lets agents interact with websites using Chromium. It combines browser control, LLM-driven decision making, and persistent browser data for tasks such as research, testing, scraping, navigation, and autonomous web workflows.

About Hosting Browser Use

Hosting Browser Use on Railway gives you a ready-to-run Browser Use environment with Chromium and persistent storage.

This template uses the official Browser Use container image and is designed as an interactive browser automation runtime rather than a public REST API or Web UI.

After deployment, Browser Use is accessed from inside the Railway container using Railway Shell, SSH, or CLI-based container access.

A Railway volume is attached to /data so browser profiles and runtime data can persist across container restarts and redeployments.

Common Use Cases

• AI-powered browser automation
• Autonomous web research
• Browser-based data collection
• Website navigation and interaction
• Browser testing and validation
• Persistent browser sessions
• AI agent experimentation
• Running Browser Use commands inside a managed Railway runtime

Dependencies for Browser Use Hosting

• Official Browser Use container image
• Chromium browser runtime
• Railway persistent volume mounted at /data
• An LLM provider API key such as OpenAI, Anthropic, Google Gemini, or Browser Use Cloud

Getting Started After Deployment

1. Configure an LLM Provider

Browser Use needs access to an LLM before it can run AI-powered browser tasks.

Open your Browser Use service in Railway and add the API key for the provider you want to use under Variables.

Common examples include:

OPENAI_API_KEY
ANTHROPIC_API_KEY
GOOGLE_API_KEY
BROWSER_USE_API_KEY

You only need to configure the provider you intend to use.

Railway will redeploy the service after environment variables are changed.

2. Understand How This Template Runs

The official Browser Use image is primarily a CLI runtime.

It does not start a public HTTP server automatically.

This means the template does not provide:

• A Web UI
• A public REST API
• A /task HTTP endpoint
• A public browser dashboard

The runtime instead contains:

• Browser Use CLI
• Python runtime
• Chromium
• Browser automation dependencies
• LLM integration support
• Persistent browser profile storage

The template keeps the container running so Browser Use can be accessed interactively when required.

Accessing Browser Use

Railway Shell / Console

The simplest way to use the deployment is to open the service shell from Railway.

Once inside the container, Browser Use commands can be executed directly.

Example:

browser-use <<'PY'
print(page_info())
PY

Browser Use executes Python passed through standard input.

More complex Browser Use automation scripts can also be executed from inside the container.

Railway SSH

If SSH access is available for the service, connect to the running container and execute Browser Use commands there.

Conceptually:

Local Machine
     ↓
Railway SSH
     ↓
Browser Use Container
     ↓
browser-use CLI
     ↓
Chromium

This is useful for interactive testing, debugging, and manually running browser automation jobs.

Railway CLI

Railway CLI can also be used as part of the workflow when accessing or operating the deployed service from a local terminal.

The exact command depends on the Railway CLI version and the container-access features available to your project.

Once connected to the running service, use the Browser Use CLI normally.

Running a Browser Automation Task

A Browser Use task conceptually follows this flow:

Browser Use command
        ↓
LLM Provider
        ↓
Browser Use Agent
        ↓
Chromium
        ↓
Website

For example, an AI browser task could instruct an agent to:

Open Hacker News, find the top five posts, and summarize them.

Browser Use uses the configured LLM to decide how to navigate, inspect, and interact with the website.

Persistent Browser Sessions

The Railway persistent volume is mounted to:

/data

Browser profile data can be stored under:

/data/profiles/default

This allows browser state to persist across container restarts and redeployments.

Depending on the workflow, persistent browser data may include:

• Cookies
• Login sessions
• Browser profile settings
• Site preferences
• Other Chromium profile data

Persistent profiles are useful for browser tasks that need authenticated sessions or repeated access to the same websites.

Public Networking

Browser Use does not need a Railway public domain for this template.

The official runtime does not expose a ready-to-use HTTP application, so generating a public Railway domain does not automatically make Browser Use accessible through a browser or REST client.

If your application needs to call Browser Use over HTTP, an additional API wrapper is required.

Important: Runtime vs API Service

This template should be treated as a Browser Use runtime.

It provides the environment needed to run Browser Use and Chromium, but it does not convert Browser Use into a continuously listening HTTP service.

The architecture is:

Railway
└── Browser Use Runtime
    ├── Browser Use CLI
    ├── Python
    ├── Chromium
    ├── LLM integrations
    └── Persistent /data volume

It is not:

Application
    ↓ HTTP
Browser Use REST API
    ↓
Chromium

A separate REST API wrapper would be needed for that architecture.

Browser Use vs Other Browser Automation Tools

Why Deploy Browser Use on Railway?

Railway is a singular platform to deploy your infrastructure stack. Railway will host your infrastructure so you don't have to deal with configuration, while allowing you to vertically and horizontally scale it.

By deploying Browser Use on Railway, you are one step closer to supporting a complete full-stack application with minimal burden. Host your servers, databases, AI agents, and more on Railway.