Split PDF

Split PDF takes one PDF and cuts it into new ones: pull out the pages you need, remove the pages you don't, or break the file into one PDF per page or per range. It is for everyday jobs like sending only the signed page of a contract, dropping a blank scan page, separating a batch of scanned invoices, or taking one chapter out of a long report.

There are four modes. Extract pages makes one new PDF from the pages you list, in the order you type them — "5, 1-3" gives page 5 first, then 1, 2 and 3. Split every page makes a separate one-page PDF for every page, which you can download together as a ZIP. Split by ranges makes one PDF per comma-separated range, so "1-3, 4-6, 7-" on a 10-page file gives three files. Delete pages keeps everything except the pages you list. Page numbers start at 1, as in every PDF viewer, and "8-" means page 8 to the end. As soon as you choose a file, the tool reads it and shows how many pages it has, so you know which numbers are valid.

The work is done in your browser by pdf-lib, an open-source PDF library. Pages are copied into new documents exactly as they are — nothing is re-rendered or recompressed — so text stays selectable, images keep their quality, and each page keeps its size and rotation. Links that jump to another page are kept when that page is in the same output file and removed when it is not, so no link points at a page that isn't there. Each output file is named after the original and its pages, like report-pages-1-3.pdf or report-page-4.pdf.

A few things belong to the whole document and are not carried into the new files: bookmarks, the fillable form (typed values stay visible, but the fields are no longer editable), attached files, accessibility tags and digital signatures. Password-protected or permission-locked PDFs are refused with a clear message, because the library cannot decrypt them — open such a file with its password, save an unprotected copy, and split that. Files up to 200 MB are accepted.

Your PDF is never uploaded. It is read, split and handed back entirely inside your browser tab; you can check in your browser's developer tools (Network tab) that no request carries the file. The site loads its own scripts and analytics, but never your documents.

1. The file's first bytes are checked for the PDF signature, then pdf-lib opens it; encrypted and damaged files are refused with a specific message, and the page count is shown. 2. Your page list is parsed: numbers (4), ranges (2-6) and open ranges (8- for page 8 to the end), separated by commas. Every number is checked against the page count before anything is built. 3. The mode decides the output: Extract keeps the pages in typed order with repeats removed; Split every page makes one file per page; Split by ranges makes one file per range; Delete keeps every page you did not list, in the original order, and refuses to delete them all. 4. For each output, the pages are copied into a new PDF with pdf-lib's copyPages. Internal links are re-pointed at the copied pages or removed if their target page is not in that file. 5. Each file is saved with compressed object streams and named after the original and its pages; with several files you can download them all as one ZIP.