Provides tools for AI image generation, editing, and description using Google's Gemini models, supporting up to 4K resolution and multi-image reference guidance.
Click on "Deploy Server".
Wait a few minutes for the server to deploy. Once ready, it will show a "Started" state.
In the chat, type @ followed by the MCP server name and your instructions, e.g., "@Gemini Image MCP generate a 4K cinematic cyberpunk city at night with neon lights"
That's it! The server will respond to your query, and you can continue using it as needed.
Here is a step-by-step guide with screenshots.
A Model Context Protocol (MCP) server for AI image generation using Google's Gemini 3.0 models (Nano Banana Pro).
generate_image - Create images from text prompts with up to 4K resolution
edit_image - Modify existing images with instructions, style transfer, multi-image mixing
describe_image - Analyze and describe images
Reference Images - Support for up to 14 reference images for style/content guidance
Related MCP server: Nano Banana MCP Server
npm install
npm run buildclaude mcp add gemini-image --env GEMINI_API_KEY=your_key -- node /path/to/dist/index.jsAdd to ~/Library/Application Support/Claude/claude_desktop_config.json:
{
"mcpServers": {
"gemini-image": {
"command": "node",
"args": ["/path/to/dist/index.js"],
"env": {
"GEMINI_API_KEY": "your_api_key_here"
}
}
}
}Copy and paste this prompt to your Zed AI assistant:
Install the gemini-image MCP server for me. Here's what you need to do:
1. Clone the repo and build it:
git clone https://github.com/dfeirstein/gemini-image-mcp.git ~/.config/zed/mcp-servers/gemini-image-mcp
cd ~/.config/zed/mcp-servers/gemini-image-mcp
npm install
npm run build
2. Add this to my Zed settings.json under "context_servers":
{
"context_servers": {
"gemini-image": {
"command": {
"path": "node",
"args": ["~/.config/zed/mcp-servers/gemini-image-mcp/dist/index.js"],
"env": {
"GEMINI_API_KEY": "YOUR_API_KEY_HERE"
}
}
}
}
}
3. Remind me to replace YOUR_API_KEY_HERE with my actual Gemini API key from https://aistudio.google.com/apikey
The MCP provides three tools: generate_image, edit_image, and describe_image using Google's Gemini 3.0 (Nano Banana Pro) models.Get a Gemini API key from Google AI Studio.
Note: Gemini 3 Pro Image (Nano Banana Pro) requires billing enabled - no free tier.
Generate a cyberpunk cityscape at night with neon lightsWith high resolution:
Generate a detailed landscape at 4K resolutionAdd sunglasses to this photo
Remove the background
Apply a watercolor styleWhat objects are in this image?
Describe the mood and compositionModel | ID | Description |
Nano Banana Pro |
| Best quality, up to 4K resolution (default) |
Flash |
| Faster, good for descriptions |
Legacy |
| Fallback option |
1K - Standard resolution (default)
2K - High resolution
4K - Ultra high resolution
1:1 - Square (default)
3:4 - Portrait
4:3 - Landscape
9:16 - Tall portrait
16:9 - Widescreen
MIT
Analyze and describe one or more images using Google Gemini. Returns text description only.
Inputs: images (required), prompt, model.
Edit one or more images using Google Gemini 3.0 (Nano Banana Pro). Supports style transfer, object manipulation, and multi-image mixing.
Inputs: prompt (required), images (required), model, imageSize, +1 more.
Generate an image using Google Gemini 3.0 (Nano Banana Pro). Supports up to 4K resolution and up to 14 reference images for style/content guidance.
Inputs: prompt (required), model, aspectRatio, imageSize, +2 more.
Recent tool additions, removals, and schema changes observed during successful MCP inspections.
Loading tool schema changelog…
Scored across 3 tools
Each tool has a clearly distinct purpose: describe_image analyzes images, edit_image modifies existing images, and generate_image creates new images. There is no overlap in functionality, making it easy for an agent to select the correct tool.
All tools follow a consistent verb_noun pattern (describe_image, edit_image, generate_image) with the same noun and clear action verbs. The naming is predictable and readable throughout.
With 3 tools, this server is well-scoped for image processing with Gemini. Each tool earns its place by covering a distinct aspect: analysis, editing, and generation, which is appropriate for the domain.
The tool set provides complete coverage for the image processing domain with Gemini: describe for analysis, edit for modification, and generate for creation. There are no obvious gaps, and agents can handle core workflows without dead ends.
MCP server for Qwen Image 3 AI image generation
MCP server for Google Veo AI video generation
MCP server for Midjourney AI image generation and editing
MCP server for Flux AI image generation