vllm-omni Model Context Protocol (MCP) Server

A framework for efficient model inference with omni-modality models

6,000 stars

Python 1 file · ~13,548 tokens #audio-generation #diffusion #image-generation #inference #model-serving #multimodal #pytorch #transformer

vllm-omni MCP Server

MCP server for vllm-project/vllm-omni

Ready-to-Use Client Configurations

{
  "mcpServers": {
    "vllm-omni": {
      "command": "npx",
      "args": [
        "-y",
        "@vllm-project/vllm-omni"
      ]
    }
  }
}

{
  "mcpServers": {
    "vllm-omni": {
      "command": "npx",
      "args": [
        "-y",
        "@vllm-project/vllm-omni"
      ]
    }
  }
}

{
  "mcpServers": {
    "vllm-omni": {
      "command": "npx",
      "args": [
        "-y",
        "@vllm-project/vllm-omni"
      ]
    }
  }
}

{
  "mcpServers": {
    "vllm-omni": {
      "command": "npx",
      "args": [
        "-y",
        "@vllm-project/vllm-omni"
      ],
      "disabled": false,
      "autoApprove": []
    }
  }
}

{
  "mcpServers": {
    "vllm-omni": {
      "command": "npx",
      "args": [
        "-y",
        "@vllm-project/vllm-omni"
      ]
    }
  }
}

docker run -i --rm mcp/vllm-omni