Web data API that turns pages into Markdown or JSON for AI models, with search, crawl and natural-language page interaction.

MinerU overview and key facts (October 2026)
Open-source document parser from OpenDataLab at Shanghai AI Laboratory: turns PDFs, scans and Office files into Markdown or JSON for AI.
What is MinerU?
MinerU is an open-source toolkit that converts PDFs, scanned images and Office documents into Markdown, JSON and other formats that language models can use. It suits teams building RAG or agent systems on large, mixed document collections.
MinerU is developed by OpenDataLab, which operates under the Shanghai AI Laboratory. Version 4.0 bundles parsing, a local document library and service tools. It accepts PDF, images, DOC/DOCX, PPT/PPTX, XLS/XLSX, RTF, OpenDocument, EPUB, OFD, HTML and CSV, and renders to Markdown, HTML, LaTeX, DOCX, EPUB, PDF and structured JSON. It offers a Python SDK, an API, batch conversion, a router for several services and a web interface. As of September 28, 2026 the repository had about 80,800 GitHub stars. An online service with API is available at mineru.net.
MinerU AI features and use cases
- Parses difficult layoutsThe Standard and Advanced tiers use models for layout analysis and a visual language model for demanding pages. Multi-column reports, tables and formulas come out as structured text that a RAG system can index.
- Reads scans and photosThe Basic tier and above apply OCR and model-based parsing to scanned or photographed documents. Paper-based records become searchable without manual typing.
- Feeds agents with citable sourcesThe document library caches results, lets agents search content and continue page by page, and keeps stable citation locators. Answers can point back to the exact block in the source.
What sets MinerU apart
- Documents are not uploaded to the official service automatically; remote parsing is an explicit choice.
- Supports OFD and other formats common in Chinese public administration alongside Western Office formats.
- Model backends are configurable, including llama.cpp, vLLM and LMDeploy for the visual language model.
Where MinerU is strongest
Developed by OpenDataLab at Shanghai AI Laboratory; it also supports OFD, the Chinese standard format for electronic documents.
Questions to ask before you buy MinerU
Read the MinerU Open Source License: it is based on Apache 2.0 but adds conditions, so have legal review it before commercial use. If you use the hosted service at mineru.net, clarify where documents are processed and on what terms, since we found no prices or data terms on its homepage in September 2026. Self-hosting needs Python 3.10 to 3.14 and GPU capacity for the higher tiers.
MinerU reviews
No reviews yet. Be the first to share how your team uses it.
Every review shows its verification level. We read each one before it goes live. How we check reviews
Questions about MinerU
Is MinerU open source?
Yes, under the MinerU Open Source License, which is based on Apache 2.0 with additional conditions.
Does MinerU upload my documents?
Not automatically. According to the project, remote parsing through the official service requires an explicit choice.
Which output formats are available?
Markdown, HTML, LaTeX, DOCX, EPUB, PDF and structured JSON content lists, depending on the tool used.
MinerU update historyno updates yet
No changes recorded for MinerU yet.
Changes the vendor made and confirmed user reports, each checked against the vendor's own pages. We review every tool monthly.
Alternatives to MinerU
Tools in the same category (Document parsers) with the most overlapping markets.
Document parsing API from LlamaIndex that prepares more than 90 formats for AI with layout awareness, with free credits and an EU region.
AWS OCR and document analysis service that extracts text, handwriting, tables, forms and expense data from scans.
Paris-based OCR and extraction API that turns invoices, receipts, IDs, resumes and custom document types into structured JSON.



