Pdf4Vllm

Overview

Pdf4Vllm is a Model Context Protocol server that reads PDF files for vision language models with intelligent text handling. It automatically detects text corruption in PDFs and seamlessly switches to image-based processing when needed, ensuring reliable document understanding regardless of PDF quality or encoding issues.

The server runs locally on your machine via the Python package pdf4vllm-mcp, installed with uvx pdf4vllm-mcp. It communicates with MCP clients over stdio and requires no authentication or external credentials. Version v1.0.2 is the current release, with source code available at PyJudge/pdf4vllm-mcp on GitHub.

Pdf4Vllm is categorized as a File System server on mcp.site and integrates with all MCP-compatible clients including Claude Desktop, Claude Code, Cursor, VS Code, Windsurf, and Zed. If you maintain this server, claim this listing to earn the Verified badge and ensure the information stays accurate.

Installation

// claude_desktop_config.json
{
  "mcpServers": {
    "pdf4vllm": {
      "command": "uvx",
      "args": [
        "pdf4vllm-mcp"
      ]
    }
  }
}