Go to Knowledge → Add & Sources → Add Knowledge → Files and drop in PDF or DOCX files. Text is extracted in your browser, so a large file never has to be uploaded twice.
How they are stored
- Each file becomes one source you can remove again as a unit.
- PDFs are split per page, so an answer can point at the right part of a long document.
- Re-uploading a file with the same name replaces the old version rather than duplicating it.
Scanned documents
A PDF that is a picture of text has no text to extract. The plugin detects this and offers optical character recognition, loaded on demand. It is slower and not perfect — a real text PDF is always better.