Product

Extract PDF data

Use POST /v1/pdf/data to extract fields that match a JSON Schema, with page citations. Omit schema for Markdown per page. Uses the text layer; ocr or an empty layer rasters up to 8 pages for vision. This is not a searchable-text OCR overlay. Billed at the AI rate.

PDF + optional JSON Schema → structured JSON. Omit schema for Markdown. Failed extracts are not billed.

Request JSON

{
  "url": "https://example.com/invoice.pdf",
  "schema": {
    "type": "object",
    "properties": {
      "vendor": { "type": "string" },
      "total": { "type": "number" }
    },
    "required": ["total"]
  }
}

Back to PDF tools →Endpoint docs →Compare alternatives →