Google Gemini

MCP Server

Google Gemini AI model integration via MCP. Access Gemini Pro, Gemini Ultra, and Gemini's multimodal capabilities.

by Google

★★★★☆ 4.2/5 (80 reviews)
Updated 2026-03-01
typescriptgoogleaimultimodalactive

Installation

npx -y gemini-mcp-server

Quick Start

```json
{
  "mcpServers": {
    "gemini": {
      "command": "npx",
      "args": ["-y", "gemini-mcp-server"],
      "env": { "GOOGLE_API_KEY": "your-key" }
    }
  }
}
```

Tools & Capabilities

generate_text
analyze_image
process_audio
embed_content
count_tokens

Key Features

  • Access to Gemini Pro, Flash, and Ultra model tiers
  • Native multimodal support: text, images, audio, video
  • Long context window up to 1 million tokens
  • Built-in code execution and structured output
  • Google Search grounding for real-time information

Use Cases

  • Generate and analyze text with Google-scale AI
  • Process and understand images and documents
  • Handle long-form content with million-token context
  • Build multimodal AI applications

Compatibility

Claude Desktop Claude Code Continue Cursor

About Google Gemini

The Google Gemini MCP server provides AI agents with access to Google's Gemini family of models. This includes Gemini Pro for text tasks, Gemini Ultra for advanced reasoning, and Gemini's native multimodal capabilities for processing images, audio, and video alongside text.

View on GitHub →

Official Website →

Language: TypeScript

Category: AI Platforms

Browse AI Platforms servers →

← Back to MCP Hub Directory
This site uses cookies from Google for advertising and analytics. Learn more