Docling

Health v2 · maintenance only

Adoption not part of the score

65603 stars · 4697 forks observed · 2026-08-28

What it is AI-extracted, prompt v1, taxonomy v1, 2026-08-29, confidence not recorded

Docling is a Python library that parses and converts documents across many formats (PDF, DOCX, PPTX, XLSX, HTML, images, audio, and more) into structured representations like Markdown and JSON, with advanced PDF understanding including layout, tables, and reading order. It integrates with the generative AI ecosystem to prepare documents for RAG pipelines and LLM consumption.

Use cases

• convert pdfs to markdown for llm ingestion
• extract tables from pdf documents
• parse docx and pptx files into structured
• prepare documents for rag pipelines
• ocr scanned pdf documents
• batch convert office documents to text
• extract document structure and reading order

When to choose

• you need high-quality PDF parsing with layout and table understanding
• you want a unified parser for many document formats in Python
• you are building RAG or gen AI pipelines that ingest documents
• you need local, self-hosted document conversion without cloud APIs

When to avoid

• you only need simple plain-text extraction with no structure
• you need a lightweight dependency-free converter
• your documents are already clean structured data like CSV or JSON

Facets

library · maturity active

parser pdf ocr nlp machine-learning rag artificial-intelligence pdf files large-language-models python cli cross-platform document-parsing document-conversion pdf-to-markdown genai table-extraction docx pptx xlsx html-parsing natural-language-processing retrieval-augmented-generation

Member repositories

For agents

markdown · JSON · MCP: product_card(name="docling-project/docling")