Docling converts PDFs, Office files and HTML into a single document object: a tree of text, tables and pictures with reading order and page provenance kept. It exports to Markdown or JSON, chunks for retrieval, and runs on your machine.
Reach for it when
- You need tables out of a PDF with their structure intact
- You are building retrieval and the parsing is the weak link
- You want it local, with no document leaving your machine
Look elsewhere when
- A simple text extract is all you need. Use pypdf
- You want a hosted service to do it. That is LlamaParse
What the lessons will cover
- 01Converting a first document
- 02The document tree
- 03Text, tables and groups
- 04Labels and reading order
- 05PDF pipelines and OCR
- 06Exporting Markdown and JSON
- 07Chunking for retrieval
- 08The command line
- 09Serializers
Others in document parsing
Common questions
Is Docling free to use?
Docling is open source and free to run yourself. You still pay whichever model provider you point it at, and this tutorial is free with no signup.
Do I need to know Python to use Docling?
Basic Python is enough. Functions, dictionaries and imports cover most of what Docling asks of you.
When should I not use Docling?
A simple text extract is all you need. Use pypdf. You want a hosted service to do it. That is LlamaParse.
Lessons for Docling are being written. Meanwhile the Agent Governance Toolkit tutorial covers the same ground: state, tools, loops, memory and human approval. Most of it carries straight over.
Start the Agent Governance Toolkit tutorial →