opendatalab/MinerU

Transforms complex documents like PDFs and Office docs into LLM-ready markdown/JSON for your Agentic workflows.

What it solves

MinerU is a high-accuracy document parsing engine designed to convert complex documents—including PDFs, DOCX, PPTX, XLSX, images, and web pages—into structured Markdown or JSON. It solves the problem of extracting high-quality, structured data from documents with complex layouts, such as multi-column layouts, scanned documents, handwriting, and cross-page tables, which is essential for LLM, RAG, and Agent workflows.

How it works

The engine utilizes a dual-engine approach combining Vision Language Models (VLM) and Optical Character Recognition (OCR) to support 109 languages. It offers three inference backends:

  • pipeline: A fast, stable, CPU/GPU-compatible backend that avoids hallucinations.
  • vlm-engine: A high-accuracy backend supporting the vLLM, LMDeploy, and mlx ecosystems.
  • hybrid-engine: A balance of high accuracy and native text extraction to minimize hallucinations.

Who it’s for

  • Developers building RAG frameworks or AI agents who need clean, structured data from various document formats.
  • Enterprise users requiring private, fully offline deployment on various AI chips (e.g., Ascend, Moore Threads).
  • No-code users who prefer using a web version, desktop client, or Gradio WebUI.

Highlights

  • Multi-format support: Native parsing for PDF, DOCX, PPTX, XLSX, and images.
  • Complex layout handling: Accurate reconstruction of formulas (to LaTeX), tables (to HTML), and human reading order with automatic header/footer removal.
  • Broad integration: Native integration with LangChain, LlamaIndex, RAGFlow, Dify, and FastGPT, as well as MCP server support for Cursor and Claude Desktop.
  • Scalable infrastructure: Supports multi-threaded concurrent inference, multi-GPU deployment via mineru-router, and a sliding-window mechanism for ultra-long documents.

Related

  • Project
    datalab-to/marker

    A high-accuracy document conversion tool that transforms PDFs and other files into structured Markdown, JSON, or HTML, specializing in tables and equations.

  • Project
    docling-project/docling

    A document parsing tool that converts complex PDFs, Office files, and multimodal content into structured Markdown or JSON for use in generative AI and RAG pipelines.

  • Project
    pymupdf/PyMuPDF

    A high-performance Python library for extracting and manipulating PDF and Office documents, featuring specialized tools for converting documents into LLM-ready Markdown for RAG pipelines.

  • Project
    pymupdf/pymupdf4llm

    A lightweight extension for PyMuPDF that converts PDFs and other documents into structured Markdown, JSON, and plain text optimized for RAG pipelines and LLM ingestion.

  • Project
    magicrew/doc7

    A document conversion tool that turns PDFs, Office files, and images into AI-ready Markdown using multimodal vision models, supporting local inference for privacy and cost-efficiency.