Nano Banana MCP
MCP server for Google's Nano Banana image generation models (Gemini). Generate and edit images via natural language.
README
@maheidem/nanobanana-mcp
MCP server for Google's Nano Banana image generation models (Gemini). Generate and edit images via natural language from Claude Code or any MCP client.
Features
- Text-to-image generation with configurable model, aspect ratio, and resolution (up to 4K)
- Image editing via natural language — add/remove elements, style transfer, inpainting
- Iterative refinement — keep editing the last generated image
- 3 model tiers — Flash (cheap), Pro (studio-quality), Flash 2 (best value)
- File paths only — no base64 in responses, prevents context window overflow
- Persistent model preference — choose once, used everywhere
Models
| Model | ID | Status | Max Resolution | Price/Image |
|---|---|---|---|---|
| Nano Banana | gemini-2.5-flash-image |
GA | 1K | $0.039 |
| Nano Banana Pro | gemini-3-pro-image-preview |
Preview | 4K | $0.134–$0.24 |
| Nano Banana 2 | gemini-3.1-flash-image-preview |
Preview | 4K | $0.045–$0.151 |
Prerequisites
- Node.js >= 18
- A Google AI API key with billing enabled (no free tier for image generation)
- Get your key at Google AI Studio
Install
Claude Code (recommended)
claude mcp add nano-banana -e GEMINI_API_KEY=your-key -- npx --yes @maheidem/nanobanana-mcp
Manual .mcp.json
{
"mcpServers": {
"nano-banana": {
"command": "npx",
"args": ["@maheidem/nanobanana-mcp"],
"env": {
"GEMINI_API_KEY": "your-key"
}
}
}
}
Tools
nb_generate_image
Generate a new image from a text prompt.
| Parameter | Required | Description |
|---|---|---|
prompt |
Yes | Text description of the image |
model |
No | Model ID (uses saved preference) |
aspectRatio |
No | e.g. 16:9, 1:1, 9:16 |
imageSize |
No | 512, 1K, 2K, or 4K |
outputDir |
No | Save directory (default: NB_OUTPUT_DIR or cwd) |
nb_edit_image
Edit an existing image with natural language instructions.
| Parameter | Required | Description |
|---|---|---|
imagePath |
Yes | Path to the source image |
prompt |
Yes | Edit instructions |
referenceImagePaths |
No | Reference images for style transfer |
model |
No | Model ID |
aspectRatio |
No | Output aspect ratio |
imageSize |
No | Output resolution |
outputDir |
No | Save directory |
nb_continue_editing
Continue editing the last generated/edited image.
| Parameter | Required | Description |
|---|---|---|
prompt |
Yes | New edit instructions |
model |
No | Model ID |
imageSize |
No | Output resolution |
outputDir |
No | Save directory |
nb_get_last_image
Get metadata about the last generated/edited image. No API call.
nb_list_models
List available models and set a preferred model (persisted to disk).
| Parameter | Required | Description |
|---|---|---|
setModel |
No | Model ID to save as default |
Environment Variables
| Variable | Required | Description |
|---|---|---|
GEMINI_API_KEY |
Yes | Google AI API key (or GOOGLE_API_KEY) |
NB_OUTPUT_DIR |
No | Default output directory for generated images |
License
MIT
Recommended Servers
playwright-mcp
A Model Context Protocol server that enables LLMs to interact with web pages through structured accessibility snapshots without requiring vision models or screenshots.
Magic Component Platform (MCP)
An AI-powered tool that generates modern UI components from natural language descriptions, integrating with popular IDEs to streamline UI development workflow.
Audiense Insights MCP Server
Enables interaction with Audiense Insights accounts via the Model Context Protocol, facilitating the extraction and analysis of marketing insights and audience data including demographics, behavior, and influencer engagement.
VeyraX MCP
Single MCP tool to connect all your favorite tools: Gmail, Calendar and 40 more.
graphlit-mcp-server
The Model Context Protocol (MCP) Server enables integration between MCP clients and the Graphlit service. Ingest anything from Slack to Gmail to podcast feeds, in addition to web crawling, into a Graphlit project - and then retrieve relevant contents from the MCP client.
Kagi MCP Server
An MCP server that integrates Kagi search capabilities with Claude AI, enabling Claude to perform real-time web searches when answering questions that require up-to-date information.
E2B
Using MCP to run code via e2b.
Neon Database
MCP server for interacting with Neon Management API and databases
Exa Search
A Model Context Protocol (MCP) server lets AI assistants like Claude use the Exa AI Search API for web searches. This setup allows AI models to get real-time web information in a safe and controlled way.
Qdrant Server
This repository is an example of how to create a MCP server for Qdrant, a vector search engine.