PDF to JSON
Extract structured data from PDFs
Each field gets a type, such as text, number, date, yes or no, or a list, plus a short description that tells the AI what to look for. Values must appear in the document: anything not stated is left empty instead of guessed, and you can correct values beside their pages before exporting. The field list is saved in this browser, ready for your next batch. It runs on an on-device model or your own API key, and with a key the text is sent to that provider.
- Choose one or more PDFs.
- Define your fields, or switch to pasting a JSON Schema, then select Extract data.
- Review the values and download JSON, a CSV row per file, or the schema.