# opendatalab/MinerU

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

Repository: https://github.com/opendatalab/MinerU
Canonical: https://ross.abutalabs.com/products/mineru
Homepage: https://opendatalab.github.io/MinerU/
Language: Python
License: NOASSERTION
License Family: other
Topics: extract-data, layout-analysis, ocr, parser, pdf, pdf-converter, python, document-analysis, pdf-parser, pdf-extractor-llm, pdf-extractor-pretrain, pdf-extractor-rag, ai4science, docx, pptx, xlsx
Last push: 2026-08-25T20:03:39+00:00

## Health v2 (maintenance only)
Score: 88/100 (v2, computed 2026-09-03T02:39:23.370411+00:00)
- activity 99, release rhythm 86, longevity 65
- inputs: {"age_days": 916, "days_push": 8, "days_rel": 19, "gap_med": 2.0, "n_releases_24m": 105}
- flags: no_license
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 78560, forks 6604 (observed 2026-08-28T04:12:21.669896+00:00)

## What it is
MinerU is a document parsing tool that converts PDFs, images, DOCX, PPTX, and XLSX files into machine-readable Markdown and JSON. It handles layout analysis, OCR in 109 languages, formula-to-LaTeX conversion, and table extraction, making documents ready for LLM and RAG pipelines.

## Use cases
- convert pdfs to markdown for llm training
- extract text and tables from scanned pdfs
- prepare documents for rag pipelines
- parse research papers with formulas into latex
- convert docx and pptx files to
- ocr documents in multiple languages
- extract structured data from complex pdf layouts

## When to choose
- you need LLM-ready markdown or JSON from complex documents
- documents contain formulas, tables, or multi-column layouts
- you need OCR for scanned or garbled PDFs
- you want a CLI, API, or WebUI for batch document conversion

## When to avoid
- you only need simple text extraction from plain digital PDFs
- you need pixel-perfect visual fidelity of the original document
- you cannot run GPU-accelerated models locally and need lightweight processing

## Facets
- artifact type: library
- maturity: active
- function: ocr, parser, pdf, nlp, rag, machine-learning
- domain: pdf, large-language-models, developer-tools
- platform: python, cli, cross-platform
- tags: document-parsing, pdf-to-markdown, layout-analysis, llm-ready, document-understanding, formula-recognition, table-extraction, natural-language-processing, data-engineering, web-server, gpu

## Member repositories
- opendatalab/MinerU (main) score 88

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:12:21.669896+00:00.
- Health v2: computed from the inputs above; adoption is never an input.
- Inferred fields (summary, facets, guidance): AI-extracted, prompt v1, taxonomy v1, on 2026-08-29T16:14:11.374000+00:00, confidence not recorded.
  - readme: https://github.com/opendatalab/MinerU (fetched 2026-08-28T04:12:21.669896+00:00, sha 653cb63df358)
  - homepage: https://opendatalab.github.io/MinerU/ (fetched 2026-08-28T17:42:27.073580+00:00, sha 4c1da0434207)
  - registry_pypi: https://pypi.org/pypi/mineru/json (fetched 2026-08-28T17:42:27.082561+00:00, sha 872683b6a784)
- Data as of 2026-08-30T08:39:29.467469+00:00.