Extraction agent
Pull pages, text, tables and form answers into new files.
Pulls content out of your documents into new files without changing them.
How to use it
- Open Extraction agent and set up the AI if you haven’t.
- Describe what to pull out, such as the pricing table into a spreadsheet and the signature page into its own PDF, and choose the documents.
- Select Plan steps and approve each step. New PDFs are added to My documents; other files download from the step, and from Agent history later.
Output
Depending on your request: chosen pages as a new PDF, one PDF per range, the text as a text or Markdown file, tables as an Excel workbook, form answers as JSON, a Word copy, or a cited summary.
Finding the right pages
To choose pages, the agent reads the start of each page. Check the pages in each step’s preview, and edit them if they’re wrong. Tables are found from how the text lines up, as in PDF to Excel. Scanned pages need OCR PDF first.
Keeping the files
Files up to 5 MB are kept in agent history, encrypted, so you can download them again later. Larger files are listed but not kept: download them when the step finishes.
Scanned pages
Scanned pages, including copier scans whose text layer is weak or garbled, are read with the quick reader first (the AI tools run OCR PDF for you), so you get a result straight away. On computers that can run it, superPDF then offers high-accuracy reading beside the result, and improves those pages in the background: in Chat and Ask multiple PDFs the pages behind your latest question come first and new questions use the improved text; elsewhere, Run again with the improved text redoes the result. It runs on this device: nothing leaves it, and it works offline once downloaded. Download: 871 MB, once, only when you choose it; delete it any time in On-device AI models. Speed: about 10 to 30 seconds a page on a recent computer (the offer shows the time for your pages); up to 20 pages are read at a time unless you choose more or pick pages. It reads English and other European languages best; check results in other scripts, and like any OCR it can misread faded or handwritten text. Running page headers, footers and page numbers are left out of the text. If anything goes wrong, the quick result stays, with a short note. See High-accuracy reading.
Limits
superPDF sets no file size or page limit.
Needs AI: download an on-device model in On-device AI models, or connect your own API key in Chat with PDF. AI can be wrong, so check the result before you rely on it.
Privacy
Document operations run in your browser. Planning uses the AI you choose: with a model on this device, nothing leaves it; with your own API key, your request, the document names and some page text go from your browser straight to your provider, never through superPDF. See Privacy: AI features.