Hyperbrowser
Cloud browsers, sandboxes and web data APIs for AI agents, plus managed agents that run browser tasks from a prompt
Hyperbrowser is infrastructure that gives AI agents a cloud browser, compute and web data. Developers connect their scripts to hosted Chrome sessions, run code in isolated sandboxes, or send a prompt to a managed agent that completes the task in a browser and returns the result.
Key features:
- Cloud-hosted Chrome controlled through Playwright or Puppeteer.
- Managed agents including HyperAgent, Browser Use and computer-use models.
- Isolated sandbox VMs with memory snapshots for reuse between runs.
- Web API to fetch pages, crawl sites or search, returning markdown, HTML, screenshots or schema-matched JSON.
- Browser profiles that keep cookies and logins between runs.
- Live session viewing and recorded replays for debugging.
Hyperbrowser is accessed through Node.js and Python SDKs, a CLI and MCP. Plans share a credit pool across browsers, sandboxes, web data and agents: the Free plan includes 5,000 credits and one concurrent session, Startup is $30/month with 30,000 credits and 25 sessions, and Scale is $100/month with 100,000 credits and 100 sessions. Enterprise is custom.