Unstructured partitions a document into elements, each with a type and metadata. One call handles PDFs, Word, HTML, email and more, and the same elements feed cleaning, chunking and whatever comes next in your pipeline.
Reach for it when
- You ingest many file types and want one code path
- You want elements with types rather than a wall of text
- You need cleaning and chunking in the same library
Look elsewhere when
- You only handle PDFs and want the richest structure. Look at Docling
- You want the hosted API rather than the local library
What the lessons will cover
- 01partition and your first elements
- 02Element types
- 03Metadata
- 04Picking a partitioner
- 05PDF strategies
- 06Cleaning element text
- 07Extracting emails and dates
- 08Chunking
- 09A pipeline end to end
Others in document parsing
Common questions
Is Unstructured free to use?
Unstructured 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 Unstructured?
Basic Python is enough. Functions, dictionaries and imports cover most of what Unstructured asks of you.
When should I not use Unstructured?
You only handle PDFs and want the richest structure. Look at Docling. You want the hosted API rather than the local library.
Lessons for Unstructured 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 →