stabgan/openrouter-mcp-multimodal

View on GitHub ↗

MCP server for OpenRouter — chat with 300+ LLMs (Claude, Gemini, GPT), analyze images / audio / video, generate images / speech / music / video (Veo 3.1, Sora, Seedance, Wan) from Claude Desktop, Cursor, Kiro, VS Code.

75 ★25 forksTypeScriptUpdated 19d ago

What you need to know

MCP server bridging multimodal LLMs on OpenRouter, enabling vision models to be called through tool interfaces.

Install

Set the OpenRouter API key and configure the server in your MCP client.

Usage

  • Send images and prompts to multimodal models via OpenRouter and receive the generated responses.

Key features

  • Access to many multimodal models through OpenRouter
  • One key covers a wide model catalog
  • Simple bridging to the MCP ecosystem

Best for

Developers wanting vision-capable models behind a single OpenRouter key.

Platforms: macOS · Linux · Web

Reviewed 2026-08-11

Topics

ai-agentclaudeclaude-desktopcursordockergeminiimage-generationllmmcpmcp-servermodel-context-protocolmultimodalopenroutersorasttttstypescriptveovideo-generationvision
Stars
75★
Forks
25
Language
TypeScript
License
Apache-2.0
Created
2025-03-26
Last push
2026-08-17