PDF Page Extractor
Pull selected pages out of any PDF into a new document — ranges like 1,3-5, processed locally with pdf-lib and downloaded instantly.
Reviewed 17 November 2026 · Free · No account needed
How to use
- Choose the PDF file.
- Enter the pages to keep (e.g. 1,3-5).
- The plan is validated against the real page count.
- Extract and the download starts.
Example
Input: 12-page PDF · keep 1,3-5
Output: 4-page PDF containing pages 1, 3, 4, 5
How it works
The pages you select are copied out of the PDF with pdf-lib into a new document. Page numbers refer to the document order, starting at 1.
The page spec ('1,3-5' syntax) is resolved against the document's real page count, then pdf-lib copies exactly those pages into a fresh document, which is saved and downloaded. The original file is untouched.
A 12-page PDF with 'keep 1,3-5' produces a 4-page PDF containing pages 1, 3, 4 and 5.
Edge cases to know
- Page numbers are 1-based, ranges include both ends, duplicates collapse, and the output is always in document order even if you type 5,1.
- The plan is validated up front — out-of-range, reversed or malformed specs stop before any file is produced.
- Encrypted PDFs open for reading, but extraction can fail if the document's restrictions are enforced.
Limits & compatibility
- Page numbers are 1-based; ranges include both ends.
- Encrypted PDFs load for viewing metadata but may fail extraction if restrictions are enforced.
- Annotations tied to removed pages stay with their page, not the new document.
Privacy
Extraction runs locally with pdf-lib in your browser. The file is never uploaded or stored.
Related tools
Questions about PDF Page Extractor
How do I choose which pages to extract?
Type them as a range list — for example 1,3-5 — into the pages box. Page numbers are 1-based, and ranges include both ends, so 1,3-5 on a twelve-page PDF yields a four-page document containing pages 1, 3, 4 and 5, in that order.
What if I ask for pages that do not exist?
The plan is validated against the document's real page count before anything is built, and an invalid range stops the extraction with a clear message. You get an error rather than a silently wrong PDF, so a typo cannot cost you a bad file.
Does it work with encrypted PDFs?
Partly. An encrypted PDF may load so you can view its metadata, but extraction can fail if the document's restrictions are enforced. There is no password handling here — remove the protection first with the appropriate tool, then extract the pages you need.
Are my documents uploaded?
No. Extraction runs locally with pdf-lib inside your browser, and the file is never uploaded or stored anywhere. One detail worth knowing: annotations tied to an extracted page travel with that page into the new document, rather than being left behind or stripped out.