MinerU

Parses documents and images to extract text, tables, and formulas with MLX acceleration on Apple Silicon.

python

3

GitHub Stars

python

Language

5 months ago

First Indexed

2 months ago

Catalog Refreshed

Documentation & install

Readme and setup notes from the catalogue, plus a client-ready config you can copy for your MCP host.

Installation

Add the following to your MCP client configuration file.

Configuration

View docs

{
  "mcpServers": {
    "tinkpa-mcp-mineru": {
      "command": "uvx",
      "args": [
        "--from",
        "mcp-mineru",
        "python",
        "-m",
        "mcp_mineru.server"
      ]
    }
  }
}

You run an MCP server that parses documents and images using MinerU. It can extract text, tables, and formulas from PDFs, screenshots, and scanned documents with MLX acceleration on Apple Silicon, making it easy to integrate powerful parsing into your Claude Code projects.

How to use

You connect to the MinerU MCP server through your MCP client and choose the parse tool to process PDFs or images. Use the parse options to extract text, tables, and formulas, and select a backend that balances speed and quality for your environment. You can test with documents like PDFs or common image formats to verify table preservation and LaTeX conversion for formulas.

How to install

Prerequisites you need before installing the MinerU MCP server:

Python is required to run the MinerU server component.
The Claude Code CLI (or equivalent MCP management tool) is used to register and run MCP servers.

Step-by-step commands to install and configure the MinerU MCP server for use with Claude Code projects:

claude mcp add --transport stdio --scope user mineru -- \
  uvx --from mcp-mineru python -m mcp_mineru.server

This command installs and configures the MinerU MCP server for all your Claude Code projects using uvx. It sets up a stdio-based connection so your client can communicate with the server directly in your workflow.


## Available tools

### parse\_pdf

Parses PDF and image files to extract structured content as Markdown or structured blocks, preserving tables and converting formulas to LaTeX.

### list\_backends

Checks system capabilities and returns recommended backends for balance between speed and quality.