- Use the Markdown output directly in your applications
- Classify and split multi-document files into separate documents
- Run field extraction on the parsed content
Use ADE Parse to Parse Documents
Use the API to parse data from documents. See the full API reference here. To get detailed information about the API response, go to JSON Response for Parsing.Rotation detection can be enabled upon request. To request this feature, contact [email protected].
Specify Documents to Parse
The API offers two parameters for specifying the document you want to parse:document: Specify the actual file you want to parse.document_url: Include the URL to the file that you want to parse.
Set Up Splits for Parsing
By default, the full document is parsed when you call the API. However, you can set thesplit parameter to page to parse each page of the document separately. When this is selected, the splits object in the API output contains a set of data for each page.
Example: Parse a Document with the API
This example shows how to parse a document with the API and save the Markdown output to a file. Materials:- Sample PDF: Wire Transfer Form
id. For example, the first chunk is the text ”# WIRE TRANSFER FORM”. The id for that chunk is 33335548-e7c3-40bd-898e-4f23d6c99d34.
To extract specific fields from the parsed Markdown, see Extract Data.
Run Parse with Our Libraries
Click one of the tiles below to learn how to run the API with our libraries.Python Library
Run Parse with our Python library.
TypeScript Library
Run Parse with our TypeScript library.
The legacy agentic-doc library does not support the API.
Migrate from the Legacy Parse Endpoint
If you’ve been using the legacy API endpoint (v1/tools/agentic-document-analysis), the API returns a different output format.
When migrating to the endpoint, you may need to update scripts that process the parsed output.
Key differences in the output:
- The output doesn’t include extraction data. Use the separate API for field extraction.
- The output is not wrapped in a
dataobject. - Each
chunksobject now has amarkdownattribute. - The chunk type is in the
typeattribute (legacy useschunk_type). - The chunk ID is in the
idattribute (legacy useschunk_id). - Bounding box coordinates use full names:
left,top,right,bottom(legacy usesl,t,r,b). - The output includes a
splitsobject showing how the document was split. - The output includes a
metadataobject with processing information.

