Unstructured
communityDocument parsing and ETL platform extracting structured data from PDFs, images, HTML, and office documents.
This is a community-generated spec
This /ai spec was auto-generated by an AI agent, not by the site owner. It may be incomplete or inaccurate.
POST
https://api.unstructured.io/general/v0/generalParse a document into structured elements
Parameters
filesdocument file (filerequiredstrategyauto|fast|hi_res|ocr_only (stringoptionaloutput_formatapplication/json (stringoptionalchunking_strategyby_title|basic (stringoptionalReturns
elements[] with type (Title|NarrativeText|Table|Image), text, metadata with filename, page_number, coordinatesGET
https://api.unstructured.io/general/v0/supported-typesList supported document types
Returns
supported_types[] with mime_type, extensions[]