How to extract pages from a PDF
- Choose a compatible PDF and check its page count.
- Enter comma-separated pages and hyphenated ranges, or select their numbers. Review your selection before continuing.
- Select Extract pages, open the new PDF and download it. Changing the selection clears the previous download.
Individual pages, ranges and order
For example, 1, 3, 5-8 keeps pages 1, 3, 5, 6, 7 and 8. Entering 3, 1, 3 produces pages 1 and 3: original order, without duplicates. Reversed ranges, out-of-bounds numbers and expressions such as “last” are not supported. This version creates one PDF; it does not split each page into a separate file.
What is preserved and what are the limitations?
Page content, text, fonts, images, size, crop and rotation are copied without converting pages into photographs or recompressing images. HTTP/HTTPS web links are supported. Original document bookmarks and metadata are not retained; PDF/A, accessibility tags and advanced layers are not guaranteed. Removing hidden or shared resources is not guaranteed: do not use this to anonymise or redact confidential information. Always review the result.
Detected digital signature structures are rejected, even with an empty field, because creating a new document does not preserve signature validity. A scanned signature is an image and is copied as page content. Forms, encrypted PDFs, attachments and annotations other than web links are also excluded. No passwords are requested. A PDF that opens in another reader may be rejected if damaged or unsupported.
Privacy and limits
Your PDF is processed in your browser. The file, its name and selected page numbers are not sent to our server or included in analytics. Remove PDF, reload or close the page to clear the selection. Maximum 25 MiB input, 500 pages and 100 MiB output. These are practical limits; devices with limited memory may need smaller files.
Need to combine documents? Merge PDF.
pdf-lib 1.17.1 · MIT licence and notices (new tab).