openrouter-mcp-multimodal
by stabgan·★ 39·Score 47
MCP server for OpenRouter providing text, image, audio, and video analysis and generation with 300+ LLMs.
Overview
This MCP server offers comprehensive multimodal capabilities through OpenRouter, connecting Claude Desktop, Cursor, Kiro, VS Code, Windsurf, or Cline to a wide range of language models. It distinguishes itself by providing both analysis and generation capabilities across multiple media types in a single package, including text processing, image analysis/generation, audio analysis/generation, and video understanding/generation. The server supports advanced features like response caching, web search, provider routing with customizable sorting, and the latest MCP 2025-06-18 specification with structured outputs and progress notifications.
Try asking AI
After installing, here are 5 things you can ask your AI assistant:
When to choose this
Choose this server when you need multimodal capabilities (text, image, audio, video) in your AI workflows and want access to OpenRouter's diverse model ecosystem through MCP.
When NOT to choose this
Avoid if you need vendor-agnostic access or if you're locked into a different LLM provider that doesn't offer OpenRouter integration.
Tools this server exposes
12 tools extracted from the READMEchat_completionSend messages to any model. Supports provider routing, model suffixes, response caching, reasoning passthrough, and web search.
analyze_imageAnalyze images from local files, URLs, or data URIs. Auto-optimized with sharp.
analyze_audioTranscribe/analyze audio from files, URLs, or data URIs.
analyze_videoAnalyze video from files, URLs, or data URIs.
generate_imageGenerate images with aspect ratio control and optional path-sandboxed disk save.
generate_audioGenerate speech or music. Auto-detects format, wraps raw PCM in WAV.
generate_videoGenerate video via async API (Veo 3.1 / Sora 2 Pro) with MCP progress notifications.
search_modelsSearch/filter models by name, provider, or modality. Paginated.
rerank_documentsRerank documents against a query (Cohere, Fireworks).
get_model_infoGet pricing, context length, and capabilities for any model.
health_checkVerify API key, OpenRouter reachability, server + protocol versions.
validate_modelCheck if a model ID exists on OpenRouter.
Comparable tools
Installation
Installation
Quick Install
npx -y @stabgan/openrouter-mcp-multimodalRequires OPENROUTER_API_KEY environment variable.
Claude Desktop Configuration
{
"mcpServers": {
"openrouter": {
"command": "npx",
"args": ["-y", "@stabgan/openrouter-mcp-multimodal"],
"env": {
"OPENROUTER_API_KEY": "sk-or-v1-..."
}
}
}
}Docker
{
"mcpServers": {
"openrouter": {
"command": "docker",
"args": [
"run", "--rm", "-i",
"-e", "OPENROUTER_API_KEY=sk-or-v1-...",
"stabgan/openrouter-mcp-multimodal:latest"
]
}
}
}Global Install
npm install -g @stabgan/openrouter-mcp-multimodalThen configure with:
{
"mcpServers": {
"openrouter": {
"command": "openrouter-multimodal",
"env": { "OPENROUTER_API_KEY": "sk-or-v1-..." }
}
}
}FAQ
- What models are supported?
- The server provides access to 300+ LLMs through OpenRouter, including models from Anthropic (Claude), Google (Gemini), OpenAI (GPT), and others. You can search and filter models by name, provider, or modality using the search_models tool.
- How do I configure provider routing?
- You can set provider preferences using environment variables like OPENROUTER_PROVIDER_SORT (price/throughput/latency), OPENROUTER_PROVIDER_IGNORE (CSV of provider slugs to exclude), and OPENROUTER_PROVIDER_DATA_COLLECTION (allow/deny). These can also be set per request in the chat_completion tool.
Compare openrouter-mcp-multimodal with
Last updated · Auto-generated from public README + GitHub signals.