They produce the same kind of result, and the only question is which is less typing.
Extract names the pages you want to keep. Right when the selection is small: pulling three pages out of a two-hundred-page report.
Delete names the pages you want gone. Right when the selection is small the other way: removing a cover sheet and two blank pages from a scan.
Both build a new document and leave your original file untouched.
How to write the page selection
Commas separate individual pages. Hyphens describe ranges.
1, 3, 5-9
That keeps page one, page three, and pages five through nine — nine pages in total, in that order.
Pages appear in the new document in the order you list them, so 5-9, 1 is a valid and meaningful instruction if you want page one at the end.
What happens to the pages
They are copied, not reprocessed. Images are not re-encoded, text is not rasterised, and nothing is compressed further.
That means the extracted pages are byte-identical in content to the originals. It also means the new file may not shrink as much as you expect: PDFs often share resources such as embedded font subsets across the whole document, and those have to come along with the pages that use them.
Why do this locally
The documents people extract pages from tend to be the ones where the other pages are the point — sending one page of a bank statement, one clause of a contract, one result from a medical report.
Uploading the whole file to a server to get one page back rather defeats the purpose. Here the document is read into memory in your own browser, the new file is built there, and nothing is transmitted. The Uploaded readout beside the result measures this page’s own network calls and reads 0 B on success.
Questions
- How do I save just one page of a PDF?
- Add the file and type that single page number. You get a new one-page document containing exactly that page, with the original left untouched on your disk.
- How do I type a page range?
- Use commas for individual pages and hyphens for ranges. "1, 3, 5-9" keeps page one, page three, and pages five through nine. Pages appear in the new document in the order you list them.
- What is the difference between extracting and deleting pages?
- They are two ways of describing the same operation, and which is easier depends on your selection. Extracting names the pages you want to keep; deleting names the pages you want gone. To keep three pages out of two hundred, extract. To lose three pages out of two hundred, delete.
- Does extracting reduce quality?
- No. Pages are copied into the new document as they are, without re-compression. Images are not re-encoded and text is not rasterised.
- Will the new file be smaller?
- Usually, but not always proportionally. A PDF may share resources such as embedded fonts across pages, so a document that keeps ten of a hundred pages will not necessarily be a tenth of the size.
- Is my document uploaded?
- No. The extraction runs in this browser tab with pdf-lib. The Uploaded readout beside the result measures the page's own outbound traffic, so a successful extraction reads 0 B.
- Will there be a watermark?
- No. Nothing is added to the pages and no feature is held behind a paid tier.
- Are password-protected PDFs supported?
- Documents that only restrict editing will usually work. Documents that require a password to open cannot be read without it and will fail. Remove the password in your PDF reader first.
- Is there a page limit?
- None imposed. There is no server and no bandwidth cost, so your device's memory is the only ceiling.
- Are form fields and bookmarks kept?
- Page content, images, links and annotations are carried across. Form fields and bookmarks belong to the document as a whole rather than to individual pages, so they may not survive.