MinerU

Provides document parsing via MinerU API with OCR, page range, and batch support.

typescript

0

GitHub Stars

typescript

Language

6 months ago

First Indexed

3 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": {
    "linxule-mineru-mcp": {
      "command": "npx",
      "args": [
        "-y",
        "mineru-mcp"
      ],
      "env": {
        "MINERU_API_KEY": "your-api-key",
        "MINERU_BASE_URL": "https://mineru.net/api/v4",
        "MINERU_DEFAULT_MODEL": "pipeline"
      }
    }
  }
}

You can run the MinerU MCP server to parse documents via its API, exposing you to fast, scalable parsing with VLM support, page ranges, batch processing, and broad language OCR coverage. This guide walks you through using the server, installing it locally, and understanding configuration and limits so you can integrate MinerU MCP into your workflow quickly and securely.

How to use

You interact with the MinerU MCP server through an MCP client or orchestration layer. Start the MCP server locally with your API key and then issue parsing requests to parse single documents or batches. You can specify model mode, target pages, OCR, and additional recognition features like formulas and tables. For batch operations, send multiple document URLs in one request and paginate the results to handle large outputs efficiently.

Core usage patterns you will likely employ:

How to install

Prerequisites: ensure you have Node.js and npm installed on your system, because the MCP server is invoked via npx and run as a JavaScript/TypeScript tool.

Install the MinerU MCP server using the provided command that wires in your API key. Run this in your terminal:

claude mcp add mineru-mcp -e MINERU_API_KEY=your-api-key -- npx mineru-mcp

Alternatively, configure the MCP server in your Claude Desktop environment using the following JSON snippet. This sets up a local stdio-based server that runs with npm/npx and passes your API key as an environment variable.

{
  "mcpServers": {
    "mineru": {
      "command": "npx",
      "args": ["-y", "mineru-mcp"],
      "env": {
        "MINERU_API_KEY": "your-api-key"
      }
    }
  }
}

Configuration overview

Configure the following environment variables to control how MinerU MCP behaves and where it talks to the MinerU service.

MINERU_API_KEY: Your MinerU API Bearer token (required)
MINERU_BASE_URL: API base URL (default: https://mineru.net/api/v4)
MINERU_DEFAULT_MODEL: Default model to use if you don’t specify one per request (default: pipeline; alternative: vlm)

Available tools

mineru_parse

Parse a single document URL with options for model, pages, OCR, formula, table recognition, language, and export formats.

mineru_status

Check a parsing task status and retrieve the download URL or concise status output.

mineru_batch

Parse multiple documents in a single batch, supporting up to 200 URLs.

mineru_batch_status

Get paginated results for a batch, with optional detail level and format.