vllm

A high-throughput and memory-efficient inference and serving engine for LLMs

88,911 stars Python 1 file · ~2,616 tokens #amd #blackwell #cuda #deepseek #deepseek-v3 #gpt #gpt-oss #inference

vllm MCP Server

MCP server for vllm-project/vllm

Ready-to-Use Client Configurations

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

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

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

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

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

docker run -i --rm mcp/vllm