Upstage Document OCR icon

Upstage Document OCR

Extract text from document images using Upstage Document OCR. Supports JPEG, PNG, BMP, PDF, TIFF, HEIC, DOCX, PPTX, XLSX, HWP, HWPX formats.

Upstage Document OCR

AI-generated

Summary

Extract text and structured data from documents using Upstage Document OCR.

Inputs

  • binaryPropertyName (required) — Name of the input item binary property that contains the file. Defaults to 'data'. Required: the item must have a binary property matching this name.
  • model — The OCR model to use. Options include 'ocr' (recommended alias for the latest stable model) and 'ocr-250904'. Defaults to 'ocr'.
  • schema — Optional parameter that specifies the response format. If set, the output is converted to the format of the corresponding OCR API. Options: 'Default (Upstage)', 'Clova', 'Google'. Defaults to empty (no conversion).
  • returnMode — Choose what data to return from the OCR response. Options: 'Full Response' (default), 'Text Only', 'Pages Array', 'Words Array', 'Confidence Score'. Defaults to 'Full Response'.

Output shape

Returns an array of n8n items. Each item's json data shape is determined by the returnMode parameter:

File size is validated against a 50MB limit. Supported input formats include JPEG, PNG, BMP, PDF, TIFF, HEIC, DOCX, PPTX, XLSX, HWP, and HWPX.

Examples

Example 1: Extract only the text content from a document image.

Set binaryPropertyName to the source binary property, leave returnMode as 'text'. Output item json will contain a 'text' property.

Example 2: Extract all words and their positions from a document.

Set binaryPropertyName to the source binary property, set returnMode to 'words'. Output item json will contain a 'words' property which is an array of word objects from all pages.

Links

Discussion