google-gemini/computer-use-preview

Health v2 · maintenance only

Flags: no_releases

Adoption not part of the score

3181 stars · 403 forks observed · 2026-08-28

What it is AI-extracted, prompt v1, taxonomy v1, 2026-08-30, confidence not recorded

A Python reference implementation of Gemini's computer-use agent that lets the model control a browser (via Playwright or Browserbase) to complete tasks from natural-language queries. It works with both the Gemini Developer API and Vertex AI.

Use cases

• automate browser tasks with an AI agent
• let Gemini control a web browser
• build a computer-use agent with Gemini API
• have an AI fill forms and navigate websites
• test browser automation driven by natural language
• run an AI agent on Vertex AI to use a computer

When to choose

• you want a ready-made agent that operates a Chrome browser from natural-language instructions
• you're already using Gemini API or Vertex AI and want to experiment with computer-use capabilities
• you need a Python starting point for building browser-controlling AI agents

When to avoid

• you need a production-grade, hardened automation solution rather than a preview/reference implementation
• you want to automate desktop applications instead of browsers
• you can't share page content and screenshots with Google's Gemini models

Facets

library · maturity active

agent-framework computer-vision llm-inference browser-extension workflow-automation artificial-intelligence large-language-models web-development python cli cross-platform gemini computer-use playwright browser-automation google vertex-ai ai-agent ai-agents automation

Member repositories

For agents

markdown · JSON · MCP: product_card(name="google-gemini/computer-use-preview")