About Unstructured
Unstructured is a document processing company and open source library that turns files into structured data for AI systems. It handles PDFs, scans, spreadsheets and many other formats.
What it does
Files are parsed into text, tables and elements that retrieval systems and agents can use. The platform applies a processing approach to each file type, so teams do not need to set that up by hand. The service is available through an API or an MCP server, which makes it easy to call from other tools.
Key features
- Parses more than 65 file types
- Extracts tables, forms and scanned text
- Output ready for retrieval and agents
- Available through an API or MCP
- Open source library on GitHub
Who it's for
Data and AI teams building retrieval pipelines from messy document collections. The site lists finance, legal, healthcare, government, insurance and manufacturing as sectors where documents are a major source of data.
Pricing and good to know
The library is open source under the Apache 2.0 licence. The hosted service has a free start option, with paid plans on the pricing page. Documents with unusual layouts or handwriting may need a quick test on a sample before you rely on the output.
Highlights
- Parse PDFs, scans and office files
- Extract tables and form fields
- Output text ready for retrieval
- Call the service through an API or MCP
- Use the open source library directly
