Why we are featuring it
Complex documents, ready for AI pipelines.
MinerU converts diverse document formats into structured Markdown or JSON while preserving order, tables, and formulas. OCR, VLM, MCP, and RAG integrations bridge files and models.
What is inside
- PDF, DOCX, PPTX, XLSX, images, and webpages.
- Markdown or JSON with LaTeX formulas and HTML tables.
- 109-language OCR and a VLM plus OCR engine.
- Python, Go, and TypeScript SDKs, CLI, REST, and Docker.