Veryfi icon

Veryfi

Extract data from invoices, receipts, checks, bank statements, W-9/W-2, IDs, and any document via Veryfi

Document → Extract Data From Any Document

AI-generated

Summary

Extracts structured data from a document using a Veryfi blueprint and returns it as JSON.

Inputs

  • fileSource (required) — Where the document comes from: 'binary' for file data from a previous node or 'url' for a public URL.
  • binaryPropertyName (required) — The name of the input binary field that contains the file (PDF, JPEG, PNG, or TIFF). Only required when fileSource is 'binary'.
  • fileUrl (required) — Public URL of the document. Only required when fileSource is 'url'.
  • fileName — Optional file name to send to Veryfi. If left empty, the file name is taken from the binary input (or omitted for URL input).
  • blueprintName (required) — Select which Veryfi blueprint to use. The list is loaded from your Veryfi account. Can be specified using an expression.
  • auto_categorize — Whether to let Veryfi auto-categorize the document. Defaults to true.
  • boost_mode — Whether to enable boost mode (returns extracted data without document image storage). Defaults to false.
  • categories — Optional list of categories to constrain the auto-categorizer.
  • external_id — Optional external identifier to associate with this document.
  • tags — Optional tags to attach to the document.

Output shape

A single JSON object containing extracted data. When confidence details are requested, fields are flattened into sibling keys (e.g., field__score, field__ocr_score).

Supports URL input only for specific operations (extractInvoice, extractReceipt, extractCheck, extractBankStatement, extractW9, extractW2, extractDriverLicense, extractPassport, extractAnyDocument). Confidence details are supported for extractInvoice, extractReceipt, extractCheck, extractBankStatement, extractW9, and extractW2.

Examples

Example 1: Extract invoice data from a binary file using a specific blueprint

fileSource = binary, binaryPropertyName = data, blueprintName = [Selected from dropdown]

Example 2: Extract data from a public invoice URL without document storage

fileSource = url, fileUrl = https://example.com/invoice.pdf, boost_mode = true

Example 3: Extract data with confidence scores flattened for easy mapping

fileSource = binary, blueprintName = [Selected from dropdown], confidenceDetails = true, flattenConfidence = true

Links

Discussion