MCP Server Gemini Image Generator

🚀 Create High-Quality Images with Gemini Image Generator MCP Server

Leverage Google's Gemini model to generate high-quality images from text prompts via the MCP protocol.

🚀 Quick Start

This MCP server enables any AI assistant to generate images using Google's Gemini AI model. It handles prompt engineering, text-to-image conversion, filename generation, and local image storage, making it easy to create and manage AI-generated images through any MCP client.

✨ Features

• Generate images from text prompts using Gemini 2.0 Flash.
• Transform existing images based on text prompts.
• Support image processing based on files and Base64 encoding.
• Automatically generate unique hash filenames to avoid duplicates.
• Save all generated and transformed images to the specified path.

📦 Installation

Requirements

• Python 3.8 or higher.
• Virtual environment management tool (e.g., venv or uv).

Dependencies

Before running server.py, you must install the following Python packages:

fastapi==0.68.1
python-multipart==0.2.1
requests>=2.25.1
Pillow>=9.0.1

🔧 Technical Details

Generate New Images

Create brand - new images from text prompts. Support various artistic styles and image sizes.

• Generate creative images by providing text descriptions.
• Support adjusting parameters such as width, height, and quality.
• Automatically generate a unique filename and save it to the specified output path.

Transform Existing Images

Transform or edit existing images:

• Apply visual effects to existing images.
• Modify attributes such as lighting and color balance.
• Add or remove image elements.
• Change the scene atmosphere and theme.

💻 Usage Examples

Basic Usage

Request format for generating new images

GeminiImageGenerator.create_image(
    prompt: str,
    width: int = 512,
    height: int = 512,
    quality: int = 90,
    style: str = "realistic",
) -> dict:

Request format for image transformation

GeminiImageGenerator.transform_image(
    image_path_or_url: str, 
    prompt: str,
    width: Optional[int] = None,
    height: Optional[int] = None,
    quality: int = 90
) -> dict:

📚 Documentation

Installation Steps

1. Clone the repository:

git clone https://github.com/your-username/gemini-image-generator.git
cd gemini-image-generator

2. Create a virtual environment and install dependencies:

python -m venv .venv
source .venv/bin/activate
pip install -e .

3. Copy the example environment file and configure the parameters:

cp .env.example .env

4. Edit the .env file to add the API key and output path:

GEMINI_API_KEY="your-gemini-api-key-here"
OUTPUT_IMAGE_PATH="/path/to/save/images"

Configure Claude Desktop

Add the following content to claude_desktop_config.json:

{
    "mcpServers": {
        "gemini-image-generator": {
            "command": "uv",
            "args": [
                "--directory",
                "/absolute/path/to/gemini-image-generator",
                "run",
                "server.py"
            ],
            "env": {
                "GEMINI_API_KEY": "GEMINI_API_KEY",
                "OUTPUT_IMAGE_PATH": "OUTPUT_IMAGE_PATH"
            }
        }
    }
}

How to Use

After installation, you can generate or transform images in the following ways:

Generate New Images

• Provide detailed text descriptions.
• Support adjusting size and quality parameters.
• Automatically generate and save to the specified path.

Transform Existing Images

• Upload a local file or provide a network URL.
• Perform intelligent editing based on prompts.
• Support cropping, rotation, and filter application.

Testing and Debugging

You can test through the FastMCP development server:

fastmcp dev server.py

After starting, you can access the MCP Inspector interface at http://localhost:5173/ to directly test the image generation tool without using Claude or other clients.

📄 License

MIT License