flashinfer Model Context Protocol (MCP) Server

FlashInfer: Kernel Library for LLM Serving

6,135 stars

Python 1 file · ~21,841 tokens #attention #cuda #distributed-inference #gpu #jit #large-large-models #llm-inference #moe

flashinfer MCP Server

MCP server for flashinfer-ai/flashinfer

Ready-to-Use Client Configurations

{
  "mcpServers": {
    "flashinfer": {
      "command": "npx",
      "args": [
        "-y",
        "@flashinfer-ai/flashinfer"
      ]
    }
  }
}

{
  "mcpServers": {
    "flashinfer": {
      "command": "npx",
      "args": [
        "-y",
        "@flashinfer-ai/flashinfer"
      ]
    }
  }
}

{
  "mcpServers": {
    "flashinfer": {
      "command": "npx",
      "args": [
        "-y",
        "@flashinfer-ai/flashinfer"
      ]
    }
  }
}

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

{
  "mcpServers": {
    "flashinfer": {
      "command": "npx",
      "args": [
        "-y",
        "@flashinfer-ai/flashinfer"
      ]
    }
  }
}

docker run -i --rm mcp/flashinfer