📦

Banana Image MCP

Name: Banana Image MCP
Rating: 3.0 (1 reviews)
Author: zengwenliang416

MCP server: Banana Image MCP

0 installs

11 stars

4 forks

Trust: 60 — Good

Installation

npx banana-image-mcp

Ask AI about Banana Image MCP

I know everything about Banana Image MCP. Ask me about installation, configuration, usage, or troubleshooting.

0/500

Loading tools...

Reviews

Documentation

Banana Image MCP

Let Claude Generate Stunning 4K Images for You
An MCP server that brings AI image generation to Claude Desktop

Quick Start · Report Bug · Request Feature

中文

About The Project

Banana Image MCP is a production-ready MCP (Model Context Protocol) server that enables Claude and other AI assistants to generate high-quality images using Google's latest Gemini image models.

Simply describe what you want, and Claude will create it — from quick concept sketches to stunning 4K professional artwork.

Why Banana Image MCP?

Zero Setup Complexity — Just add your API key and start generating
Production Ready — Built with FastMCP framework, fully tested, CI/CD enabled
Best Quality — Leverages Gemini's most advanced image models with 4K support
Smart Defaults — Intelligent model selection based on your prompts
Real-World Knowledge — Google Search grounding for accurate, factual images

(back to top)

Built With

(back to top)

Features

4K Ultra HD Output

Generate images up to 3840px with the Pro model. Perfect for professional work, marketing materials, and print-ready assets.

Dual Model Support

Flash: 2-3s, up to 1024px — for quick iterations
Pro: 5-8s, up to 4K — for final deliverables

Smart Model Selection

The server automatically picks the best model based on your prompt. Say "quick sketch" for Flash, or "4K professional" for Pro.

Google Search Grounding

Pro model uses real-world knowledge from Google Search to generate more accurate and factual images.

Flexible Aspect Ratios

Support for all common ratios: 1:1, 16:9, 9:16, 4:3, 3:2, 21:9 and more.

Natural Language Editing

Edit existing images with simple text commands like "make the sky more dramatic" or "remove the background".

(back to top)

Quick Start

Get up and running in under 2 minutes.

Prerequisites

Get a free Gemini API key from Google AI Studio
Have Claude Desktop installed

Installation

Add to your Claude Desktop config file:

{
  "mcpServers": {
    "banana-image": {
      "command": "uvx",
      "args": ["banana-image-mcp"],
      "env": {
        "GEMINI_API_KEY": "your-api-key-here"
      }
    }
  }
}

📁 Config file locations

Platform	Path
macOS	`~/Library/Application Support/Claude/claude_desktop_config.json`
Windows	`%APPDATA%\Claude\claude_desktop_config.json`
Linux	`~/.config/Claude/claude_desktop_config.json`

🔄 Updating to the latest version

When using uvx, packages are cached locally. To get the latest version:

# Clear the cache for this package
uv cache clean banana-image-mcp

# Then restart Claude Desktop

Or specify a version explicitly in your config:

"args": ["banana-image-mcp==1.0.1"]

🔌 Other MCP clients (Cursor, VS Code, Cline, etc.)

The configuration is the same for other MCP-compatible clients. Just add the server config to your client's MCP configuration file.

(back to top)

Usage

Just ask Claude to generate images naturally:

"Generate a cute cat wearing a space suit"

"Create a professional product photo of a coffee cup, 4K quality"

"Make a 16:9 YouTube thumbnail about cooking"

"Edit this image: make the sky more dramatic"

Model Comparison

Model	Speed	Max Resolution	Best For
Gemini 2.5 Flash	2-3s	1024px	Quick drafts, iterations, prototypes
Gemini 3 Pro	5-8s	4K (3840px)	Final assets, marketing, professional work

The server defaults to Pro model for best quality. Control it with keywords:

Say this...	Model Used
"quick sketch", "draft", "prototype"	Flash
"4K", "professional", "high quality"	Pro
(default)	Pro

Parameters Reference

Parameter	Type	Default	Description
`prompt`	string	required	Image description
`model_tier`	string	`"pro"`	`"flash"`, `"pro"`, or `"auto"`
`resolution`	string	`"4k"`	`"1k"`, `"2k"`, `"4k"`, `"high"`
`aspect_ratio`	string	-	`"1:1"`, `"16:9"`, `"9:16"`, `"4:3"`, `"21:9"`, etc.
`thinking_level`	string	`"high"`	`"low"` or `"high"` (Pro only)
`enable_grounding`	bool	`true`	Enable Google Search grounding (Pro only)
`n`	int	`1`	Number of images (1-4)
`negative_prompt`	string	-	What to avoid in the image

Environment Variables

Variable	Required	Default	Description
`GEMINI_API_KEY`	Yes	-	Your Gemini API key
`IMAGE_OUTPUT_DIR`	No	`~/banana-images`	Where to save generated images

(back to top)

What You Can Create

Product Photography Professional shots with studio lighting	Concept Art Fantasy landscapes, character designs	Marketing Materials Social graphics, banners, thumbnails
Technical Diagrams Flowcharts, architecture diagrams	Photo-realistic Images Portraits, nature, urban photography	UI/UX Mockups App interfaces, web designs

(back to top)

Roadmap

See the open issues for a full list of proposed features and known issues.

(back to top)

Development

# Clone the repository
git clone https://github.com/zengwenliang416/banana-image-mcp.git
cd banana-image-mcp

# Install dependencies
uv sync

# Run in development mode
fastmcp dev banana_image_mcp.server:create_app

# Run tests
pytest

# Lint and format
ruff check .
ruff format .

(back to top)

Contributing

Contributions make the open source community amazing. Any contributions you make are greatly appreciated.

Fork the Project
Create your Feature Branch (git checkout -b feature/AmazingFeature)
Commit your Changes (git commit -m 'Add some AmazingFeature')
Push to the Branch (git push origin feature/AmazingFeature)
Open a Pull Request

(back to top)

License

Distributed under the MIT License. See LICENSE for more information.

(back to top)

Contact

Wenliang Zeng - @zengwenliang416

Project Link: https://github.com/zengwenliang416/banana-image-mcp

(back to top)

Acknowledgments

FastMCP - The MCP framework that powers this server
Google Gemini - The AI models behind image generation
Anthropic MCP - The protocol specification
Best-README-Template - README inspiration

(back to top)