pollinations.ai MCP Server v2.0

A Model Context Protocol (MCP) server for pollinations.ai that enables AI assistants to generate images, videos, text, and audio.

What's New in v2.0

• New API endpoint: Uses gen.pollinations.ai - the unified pollinations.ai gateway
• Authentication: Simple API key system (pk*/sk* keys) replaces OAuth
• Video generation: New generateVideo tool with veo, seedance, seedance-pro
• Chat completions: OpenAI-compatible chatCompletion tool with function calling
• Dynamic models: Models fetched from API - always up to date, no hardcoding!
• SDK upgrade: Updated to MCP SDK 1.25.1 with latest protocol support

Quick Start

# Run directly with npx (no installation required)
npx @pollinations_ai/mcp

Or install globally:

npm install -g @pollinations_ai/mcp
pollinations-mcp

Authentication

Get your API key at pollinations.ai

Key Types:

• pk_ (Publishable): Client-safe, rate-limited (1 pollen per IP per hour)
• sk_ (Secret): Server-side only, no rate limits, can spend Pollen

Set your key via environment variable or the setApiKey tool:

# Environment variable
export POLLINATIONS_API_KEY=pk_your_key_here
npx @pollinations_ai/mcp

Available Tools

Image & Video Generation

| Tool | Description | | ------------------ | -------------------------------------------- | | generateImageUrl | Generate an image URL from text prompt | | generateImage | Generate image and return base64 data | | generateVideo | Generate video (veo, seedance, seedance-pro) | | listImageModels | List available models (dynamic) |

Image parameters:

• prompt (required): Text description
• model: flux, turbo, gptimage, kontext, seedream, seedream-pro, nanobanana, nanobanana-pro, zimage
• width, height: Image dimensions (default: 1024)
• seed: Reproducible results
• enhance: Improve prompt
• negative_prompt: What to avoid
• quality: low, medium, high, hd
• image: Reference image URL for image-to-image
• transparent: Transparent background

Video parameters:

• model: veo (text-to-video), seedance, seedance-pro (image-to-video)
• duration: Video length in seconds
• aspectRatio: 16:9, 9:16, etc.
• audio: Enable audio (veo only)

Text Generation

| Tool | Description | | ---------------- | ---------------------------------------- | | generateText | Simple text generation | | chatCompletion | OpenAI-compatible chat with tool calling | | listTextModels | List available models (dynamic) |

Text models include: openai, openai-fast, openai-large, gemini, gemini-fast, gemini-large, claude, claude-fast, claude-large, deepseek, grok, mistral, qwen-coder, and more!

Chat completion features:

• Multi-turn conversations
• Function/tool calling
• JSON response format
• Audio output (with openai-audio model)

Audio Generation

| Tool | Description | | ----------------- | ------------------------- | | respondAudio | AI responds with speech | | sayText | Text-to-speech (verbatim) | | listAudioVoices | List available voices |

Voices: alloy, echo, fable, onyx, nova, shimmer, coral, verse, ballad, ash, sage, amuch, dan

Formats: mp3, wav, flac, opus, pcm16

Authentication

| Tool | Description | | ------------- | ------------------------ | | setApiKey | Set API key for requests | | getKeyInfo | Check current key status | | clearApiKey | Remove stored key |

Claude Desktop Integration

npx @pollinations_ai/mcp install-claude-mcp

Or manually add to your Claude Desktop config:

{
  "mcpServers": {
    "pollinations": {
      "command": "npx",
      "args": ["@pollinations_ai/mcp"],
      "env": {
        "POLLINATIONS_API_KEY": "pk_your_key_here"
      }
    }
  }
}

Examples

Generate an image

Generate an image of a sunset over mountains using the flux model

Generate a video

Create a 6-second video of waves crashing on a beach using veo

Chat with function calling

Use chatCompletion to have a conversation about the weather, with the ability to call a weather API

Text-to-speech

Say "Hello, welcome to pollinations.ai!" using the nova voice

System Requirements

• Node.js: Version 18.0.0 or higher

API Reference

All requests go through https://gen.pollinations.ai

| Endpoint | Description | | --------------------------- | -------------------- | | GET /image/{prompt} | Generate image/video | | GET /text/{prompt} | Generate text | | POST /v1/chat/completions | Chat completions | | GET /image/models | List image models | | GET /text/models | List text models |

Full API docs: enter.pollinations.ai/api/docs

Migration from v1.x

Breaking Changes

1. Authentication: OAuth removed → Use API keys (pk*/sk*)
2. Removed tools: startAuth, exchangeToken, refreshToken, getDomains, updateDomains
3. New tools: generateVideo, chatCompletion, setApiKey, getKeyInfo, clearApiKey
4. API endpoint: Now uses gen.pollinations.ai

Upgrade Steps

1. Get API key from pollinations.ai
2. Update: npm update @pollinations_ai/mcp
3. Set your key: Use setApiKey tool or POLLINATIONS_API_KEY env var

License

MIT

Links

• pollinations.ai
• API Documentation
• GitHub Issues