PDF Vector icon

PDF Vector

Convert PDFs, Word, Excel documents, and images to clean markdown, extract structured data with AI, process invoices with specialized parsing, and search millions of academic papers across PubMed, ArXiv, Google Scholar, and more.

Actions18

Invoice → Extract

AI-generated

Overview

The node extracts structured data from invoices using AI-powered parsing. It supports input via a public URL or binary data from a previous node. Users provide an extraction prompt and a JSON schema defining the desired data structure. The node processes various file types including PDF, DOCX, XLSX, CSV, PNG, and JPG, and supports multiple AI model tiers for different accuracy and cost preferences. This node is useful for automating data extraction from invoices for accounting, auditing, or data entry tasks.

Use Case Examples

  1. Extracting invoice number, date, and total amount from a PDF invoice to automate bookkeeping.
  2. Parsing vendor details and line items from an image of a receipt for expense tracking.

Properties

Name Meaning
Input Type Specifies how the invoice is provided, either via a public URL or as binary data from a previous node.
Invoice URL The public URL of the invoice file, required if input type is URL.
Input Binary Field The name of the binary property containing the file, required if input type is binary data.
Model Selects the AI model tier to use for extraction, balancing accuracy and cost.
Extraction Prompt Instructions specifying what data to extract from the invoice.
JSON Schema Defines the structure of the extracted data in JSON format.
Document ID Optional identifier for usage tracking, returned in the response.

Output

JSON

  • title - Extracted title from the invoice as defined by the prompt and schema.
  • summary - Extracted summary or additional structured data from the invoice.

Dependencies

  • Requires an API key credential for the PDF Vector API service.

Troubleshooting

  • Ensure the input file URL is publicly accessible if using URL input type.
  • Verify the binary property name matches the output of the previous node if using binary input.
  • The extraction prompt must be at least 4 characters long to be valid.
  • The JSON schema must be correctly formatted and match the expected output structure.
  • API errors may occur if the file size or page count exceeds model limits; select an appropriate model tier.

Links

Discussion