Create Parse
Create a parse.
Extracts the full text of a document into per-page and concatenated text using
the chosen model. Tables are rendered in the requested table_parsing_format, and
optional instructions steer the parse. Returns the stored Parse with its output
and usage, and responds with 201.
Parse a document into normalized text and persist the result as aDocumentation Index
Fetch the complete documentation index at: https://docs.retab.com/llms.txt
Use this file to discover all available pages before exploring further.
Parse resource that can later be retrieved via GET /v1/parses/{parse_id} or listed via GET /v1/parses.
Authorizations
Body
Public create-parse request body.
A file represented by its filename and a base64 data url.
- MIMEData
- FileRef
The model to use for parsing
Format used to render tables extracted from the document
markdown, yaml, html, json DPI used when rasterizing pages for the parser
x >= 72Free-form instructions appended to the system prompt to steer the parse.
If true, skip the LLM cache and force a fresh completion
If true, run asynchronously: returns immediately with status 'queued' and an empty output. Poll GET /v1//{id} until status is terminal. Mutually exclusive with stream.
Response
Successful Response
A parse result: the per-page and full-document text extracted from a document.
Unique identifier of the parse
Information about the parsed file
Model used for parsing
Format used to render tables extracted from the document
markdown, yaml, html, json DPI used when rasterizing pages for the parser
The parsed document content
Free-form instructions supplied with the parse request.
Lifecycle status. The synchronous path returns 'completed'. Background runs progress pending -> queued -> in_progress -> completed | failed | cancelled.
pending, queued, in_progress, completed, failed, cancelled Error details when a background run fails; null otherwise. Always present so consumers can read it without an existence check.
Usage information for the parse operation