Extract PDF pages
Pick the pages you need — by tapping them or by typing a range — and get them as one new PDF, or as one file per page. Private, and never stored.
- Never stored
- No queue, no waiting
- No signup, no watermark
How it works
Add your PDF
Drop the PDF onto the page. Previews of every page appear right away.
Pick your pages
Tap the pages you want, or type a range like 1-3, 9 in the box. Select all and Clear are one tap away.
Choose the output
One PDF containing your pages, or a ZIP with each page as its own file — then download.
What an extracted page brings with it
The word to take seriously in "a clean new document" is new. Your pages are copied into a document created from nothing, so the page content is exact — no recompression, no rescaling, the same fonts and the same images — while the original's document-level furniture is simply not there to inherit. No outline, no page labels, no attachments, no title or author, and no form. That last one has a visible consequence worth knowing before you send the file on: the widget for a filled-in field is drawn on the page and comes across with it, so the answers still show, but the dictionary that told readers those widgets were a form does not. An extracted form page looks complete and can no longer be filled in.
Page labels are the one that catches people out. A report whose front matter is numbered i, ii, iii carries those labels in the document, not on the pages, so extracting page three of the preface gives you a one-page PDF that every reader calls page 1 — while the numeral printed in the artwork still says iii. Nothing is wrong with the file; two different numbering systems have simply stopped agreeing.
"A separate PDF per page" multiplies everything the pages had in common, because each file has to be self-contained. A font subset embedded once in the source is embedded again in every single piece that uses a character from it, and a letterhead drawn on all forty pages is stored forty times. A ZIP of single pages routinely totals more than the document it came from, and on a long document it is built entirely in memory, which is the thing most likely to defeat a phone here.
When you want something else
Pages come out in document order here, always. When the order is the point — pulling an appendix to the front, or interleaving two scans — qpdf in.pdf --pages . 9 . 1-3 . 12-14 -- out.pdf takes ranges in whatever sequence you write them, repeats a page as often as you name it, and because the input stays the primary document it keeps the outline and the document information that an extract here cannot.
For pulling a range out of something genuinely large — a thousand-page scan, or a file bigger than a browser tab can hold — both qpdf and Poppler's pdfseparate -f 5 -l 12 in.pdf page-%d.pdf stream from disk rather than loading the document into memory first, and neither has an upper size at all.
Frequently asked questions
What is the difference between extracting and splitting?
Extracting keeps only the pages you choose and discards the rest. Splitting is about breaking one document into several. Both live here: choose "A separate PDF per page" and extracting behaves exactly like a split.
Can I type page numbers instead of tapping?
Yes. The range box accepts numbers and ranges together — "1-3, 9, 12-14" — which is far faster than tapping in a long document. The grid updates to show exactly what you asked for.
Do extracted pages keep their quality?
Yes. Pages are copied exactly as they are — no recompression, no rescaling, no watermark.
Is my PDF stored anywhere?
No. Nothing is stored at any point. The file is never stored.
Will the extracted pages be in the order I tapped them?
They come out in document order, which is what people almost always want. To put pages in a custom order, extract them first and then use Rearrange PDF pages.
Good to know: Bookmarks, links and form fields tied to pages you leave behind do not come across — the extracted file is a clean new document containing your chosen pages.
Put this tool on your website
Free for any blog, class page or help article. Paste one snippet and your visitors can use it right on your page.