Pollinations MCP Setup Guide
MCP (Model Context Protocol) lets AI assistants call tools — including image generation — during a chat. Pollinations maintains MCP packages in the community ecosystem; this guide explains the workflow.
When to use MCP vs the browser
- Browser: quick one-off images and trying prompts.
- MCP: repeated workflows inside coding or writing tools (storyboards, mockups, batch ideas).
- HTTP API: custom apps and servers you deploy yourself.
Typical setup flow
- Get API access from pollinations.ai if required for your usage tier.
- Find the current MCP package in the official Pollinations GitHub organization (search “pollinations mcp”).
- Add the server block to your client config (command, args, env vars for API keys).
- Restart the client and verify tools like image generation appear in the tool list.
Official resources
Always prefer upstream docs for breaking changes:
- pollinations.ai — platform and API keys
- GitHub — pollinations/pollinations — MCP and community projects
- This site’s API notes — curl and Python patterns
- Python API guide — simple HTTP without MCP
Frequently Asked Questions
What is Pollinations MCP?
It is a Model Context Protocol server that exposes Pollinations generation tools to compatible clients, so your assistant can create images or text without you copying URLs manually.
Do I need an API key for MCP?
Production use often requires a Pollinations API key or Pollen credits. Check official docs at pollinations.ai for current authentication.
Which apps support MCP?
Cursor, Claude Desktop, and other MCP-capable clients can load server definitions from their settings. Exact UI labels change between versions.
Related Articles
Where to Go Next
Image Generator
Create images at /image-generation/ or the homepage tool.
More guides
Browse all blog articles and the prompts library.