OpenRouter Multimodal

v4.5.1APIsstable

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.

ai-agentclaudeclaude-desktopcursordocker
Share:
40
Stars
0
Downloads
0
Weekly
0/5

What is OpenRouter Multimodal?

OpenRouter Multimodal is a Model Context Protocol (MCP) server that allows AI assistants like Claude, Cursor, and VS Code to 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, curs...

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.

This server falls under the APIs category on MCPgee, the world's largest MCP server directory with 33,000+ servers.

Features

  • MCP server for OpenRouter — chat with 300+ LLMs (Claude, Gem

Use Cases

Access 300+ LLMs
Image and audio analysis
Multimodal content generation
stabgan

Maintainer

LicenseMIT License
Languagetypescript
Versionv4.5.1
UpdatedMay 20, 2026
Statushealthy
Maintenanceactive

Works with

ClaudeOpenAIwindowsmacoslinux

Installation

NPM

npx -y @stabgan/openrouter-mcp-multimodal

Manual Installation

npx -y @stabgan/openrouter-mcp-multimodal

Configuration

Configuration Details

Config File

claude_desktop_config.json

Performance

Response Metrics

Response Time< 200ms
ThroughputMedium

Resource Usage

Memory UsageLow
CPU UsageLow

How to Set Up and Use OpenRouter Multimodal

OpenRouter MCP Multimodal Server is a TypeScript MCP server that acts as a unified gateway to over 300 AI language models and multimodal generation services via OpenRouter's API, accessible from Claude Desktop, Cursor, Kiro, and VS Code. It provides 14 tools covering text chat completion, image and audio analysis, video understanding, image generation, video generation with models like Veo 3.1 and Sora 2 Pro, audio synthesis, model search, and health checks — all within a single package secured by SSRF protection and path sandboxing.

Prerequisites

  • Node.js 18 or later installed
  • An OpenRouter API key from openrouter.ai/keys (free tier available)
  • An MCP-compatible client: Claude Desktop, Cursor, VS Code with MCP extension, or Kiro
  • Sufficient OpenRouter credits for media generation tasks (image, video, audio)
1

Get an OpenRouter API key

Register at openrouter.ai and create an API key from the Keys page. Free-tier keys work for text models; generation models require credits.

2

Confirm Node.js is available

Verify Node.js 18 or later is installed, as npx is used to run the server without a global install.

node --version
3

Configure Claude Desktop

Add the server to your Claude Desktop configuration at ~/Library/Application Support/Claude/claude_desktop_config.json.

{
  "mcpServers": {
    "openrouter-multimodal": {
      "command": "npx",
      "args": ["-y", "@stabgan/openrouter-mcp-multimodal"],
      "env": {
        "OPENROUTER_API_KEY": "sk-or-v1-your-key-here"
      }
    }
  }
}
4

Configure optional environment variables

Tune the server behavior with additional variables: set a default model, specify an output directory for saved media files, or adjust the log level.

"OPENROUTER_DEFAULT_MODEL": "anthropic/claude-sonnet-4",
"OPENROUTER_OUTPUT_DIR": "/Users/yourname/ai-output",
"OPENROUTER_LOG_LEVEL": "warn"
5

Restart Claude Desktop and run a health check

Reload Claude Desktop. All 14 OpenRouter tools should appear. Ask Claude to run a health_check to confirm the API key is valid and the server is connected.

OpenRouter Multimodal Examples

Client configuration

Full claude_desktop_config.json block for the OpenRouter MCP Multimodal server.

{
  "mcpServers": {
    "openrouter-multimodal": {
      "command": "npx",
      "args": ["-y", "@stabgan/openrouter-mcp-multimodal"],
      "env": {
        "OPENROUTER_API_KEY": "sk-or-v1-your-key-here",
        "OPENROUTER_DEFAULT_MODEL": "nvidia/nemotron-nano-12b-v2-vl:free",
        "OPENROUTER_OUTPUT_DIR": "/Users/yourname/Downloads/ai-output",
        "OPENROUTER_LOG_LEVEL": "warn"
      }
    }
  }
}

Prompts to try

Example prompts covering the range of OpenRouter multimodal capabilities.

- "Use OpenRouter to chat with meta-llama/llama-3.3-70b-instruct: summarize the theory of relativity in 3 bullet points"
- "Analyze the image at /Users/me/photo.jpg using google/gemini-2.5-pro and describe any text visible in it"
- "Search OpenRouter for the cheapest model that supports vision"
- "Generate an image of a cyberpunk cityscape at night and save it to my output folder"
- "Generate a short looping video of ocean waves at sunset using google/veo-3.1 and save as waves.mp4"
- "Using OpenRouter, rerank these three documents by relevance to the query 'machine learning inference optimization'"

Troubleshooting OpenRouter Multimodal

401 Unauthorized when calling any tool

The OPENROUTER_API_KEY must be set correctly and must start with sk-or-v1-. Double-check the value in your config has no extra characters. Regenerate the key at openrouter.ai/keys if needed.

Video or image generation tool returns an error about insufficient credits

Media generation models on OpenRouter require prepaid credits. Free-tier keys only cover free-tier text models. Add credits at openrouter.ai/credits and retry the generation request.

Generated files are not saved to disk

Set OPENROUTER_OUTPUT_DIR to an absolute path that exists and is writable. Without this variable, the server may return a URL instead of saving the file. If the target path is outside the default sandbox, also set OPENROUTER_ALLOW_UNSAFE_PATHS=1.

Frequently Asked Questions about OpenRouter Multimodal

What is OpenRouter Multimodal?

OpenRouter Multimodal is a Model Context Protocol (MCP) server that 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. It connects AI assistants to external tools and data sources through a standardized interface.

How do I install OpenRouter Multimodal?

Install via npm with the command: npx -y @stabgan/openrouter-mcp-multimodal. Then add the server configuration to your AI client's JSON config file (e.g., claude_desktop_config.json or .cursor/mcp.json).

Which AI clients work with OpenRouter Multimodal?

OpenRouter Multimodal works with all major MCP-compatible AI clients including Claude Desktop, Claude Code, Cursor, VS Code (GitHub Copilot), Windsurf, and Cline.

Is OpenRouter Multimodal free to use?

Yes, OpenRouter Multimodal is open source and available under the MIT License license. You can use it freely in both personal and commercial projects.

Browse More APIs MCP Servers

Explore all apis servers available in the MCPgee directory. Each server includes setup guides for Claude, Cursor, and VS Code.

Quick Config Preview

{ "mcpServers": { "openrouter-mcp-multimodal-server": { "command": "npx", "args": ["-y", "@stabgan/openrouter-mcp-multimodal"] } } }

Add this to your claude_desktop_config.json or .cursor/mcp.json

Read the full setup guide →

Ready to use OpenRouter Multimodal?

Browse our complete directory of 33,000+ MCP servers, read setup guides for your editor, and start building with the Model Context Protocol.

33,000+ ServersFree & Open SourceStep-by-Step Guides