Geminimcpserver

🚀 GeminiMcpServer

GeminiMcpServer is a Model Context Protocol (MCP) server that seamlessly connects LM Studio (or other MCP - compatible clients) with the Google Gemini API for image generation and multimodal task processing.

🚀 Quick Start

This server enables you to integrate LM Studio with Google Gemini API for various tasks. Follow the steps below to get started.

✨ Features

• 🔗 Full MCP Support – Real - time integration with MCP clients such as LM Studio.
• 🎨 Image Generation – Generate images using the Google Gemini 2 model.
• 🖼 Multimodal Input – (Optional) Process both text and image inputs simultaneously.
• ⚡ Hybrid Local and Cloud Workflow – Run LM Studio locally while leveraging the cloud computing power provided by the Gemini API.

📦 Installation

System Requirements

• Node.js v20
• Google Gemini API Key (You can apply for it at Google AI Studio)
• MCP - compatible client (e.g., LM Studio)

Installation Steps

git clone git@github.com:bowwowxx/GeminiMcpServer.git
cd GeminiMcpServer
npm install

Create Google API Key

1. Go to Google AI Studio.
2. Create a new API key.

Set Environment Variables

Create a .env file in the project root directory and add the following:

GEMINI_API_KEY="your_api_key_here"

💻 Usage Examples

Start the Server

npm start

Run Tests

npx tsx testapi.js

This test uses the experimental image generation API of Gemini 2 to create images.

📚 Documentation

MCP Request Example

{
  "tool": "generateImage",
  "params": {
    "prompt": "A photorealistic 3D rendered pig standing in a sunny field",
    "outputFormat": "png",
    "aspectRatio": "16:9"
  }
}

LM Studio Configuration Example

{
  "mcpServers": {
    "GeminiMcpServer": {
      "command": "npm",
      "args": [
        "run",
        "start"
      ],
      "cwd": "/Users/bowwow/github/GeminiMcpServer"
    }
  }
}