gemini-mcp and gemini-webapi-mcp
These are complements: one enables Claude to call Gemini's API programmatically, while the other exposes Gemini's web interface capabilities (image generation/editing) through an alternative authentication method, allowing different integration patterns depending on whether API access or browser-based functionality is needed.
About gemini-mcp
RLabs-Inc/gemini-mcp
MCP Server that enables Claude code to interact with Gemini
Implements Gemini 3 models (Pro/Flash) and supporting services through the Model Context Protocol, exposing 20+ tools spanning generative capabilities (images, videos, code execution) and research features (web search, YouTube/document analysis, token counting). Communicates via stdio transport and integrates seamlessly with Claude Code via the official MCP Registry, while offering a standalone CLI for direct access to all functionality. Built with async task polling for video generation and context caching for efficient repeated queries on large documents.
About gemini-webapi-mcp
AndyShaman/gemini-webapi-mcp
MCP server for Google Gemini — free image generation, editing & chat via browser cookies. No API keys needed.
Implements MCP stdio transport to expose Gemini's web API through Claude and other MCP clients, with automatic Chrome cookie extraction and support for multi-turn conversations. Features include server-side 2x upscaling of generated images, mathematically precise watermark removal via reverse alpha blending, and file analysis across video, PDF, and document formats. Supports model selection (Flash, Pro, Flash-Thinking) with configurable language output and multi-account Google authentication via environment variables.
Related comparisons
Scores updated daily from GitHub, PyPI, and npm data. How scores work