ByteScope

Split PDF

Cut a PDF into parts, pull out the pages you need, or drop the ones you don't — planned first, then done in this tab, nothing uploaded.

Your files never leave your browser — all processing is local.

Drop a PDF here

or click to browse — processed in this tab, nothing uploaded

Splitting does not make files much smaller, and there are no page previews here, so check the page numbers in your PDF reader first.

About this tool

Send pages 3 to 5 of a contract and nothing else; break a 200-page scan into 10-page chunks for a portal with an upload limit; take the blank pages out of a scanned batch. This page splits one PDF five ways: by ranges you type (1-3, 4-10, 11-, one part per range), every N pages, one page per file, extract (chosen pages into a single file, in the order you list them — 5,1,2 comes out in that order, and a page may repeat), or delete (everything except the pages you name). Nothing is uploaded; the file is read and rewritten in this tab.

You see the plan before anything is written

Type the mode and the pages, and a list appears: how many parts, which pages each one contains, and the file name each will get (report-p1-3.pdf, report-part-02.pdf, report-edited.pdf). Only when that looks right do you run it. Every part is loaded again after it is written and its page count is checked against the plan; a mismatch is an error, not a download. One part downloads directly; several are packed into a ZIP, or written straight into a folder you pick where the browser allows it.

Honest limits

This first version shows no page thumbnails, so you choose pages by number. Check the page numbers in your PDF reader first — the page count is shown, the page contents are not. Splitting is not compression: fonts and images shared across the document are carried in every part that uses them, so three parts of a 30 MB file can add up to more than 30 MB. And as with any tool that copies pages into a new document, bookmarks, cross-page links, form fields and accessibility tags may not survive, while a digital signature certainly will not.

The parts are written as fresh documents with no Document Information Dictionary — no author, producer or dates from the original, no per-page XMP, and nothing naming this tool. That is document information only: comments and other annotations are page content, so each part keeps them along with the name of whoever wrote them — delete them in your PDF reader first if they should not travel. Password-protected PDFs cannot be opened here; you are told so plainly instead of being handed an empty file, and deleting every page is refused too. Everything else is plain File API and JavaScript in this tab: turn the network off and it still works, which is what you want for contracts, transcripts and medical records — the files people most often need to split.

Frequently asked questions

Is my PDF uploaded anywhere?

No. The file is opened locally, split in JavaScript in this tab, and each part is a download the browser built in memory. There is no API behind this page, so nothing can be sent even in principle. Turn the network off and it still works.

Can I pull out specific pages in my own order?

Yes. Use the extract mode and list the pages: 5,1,2 produces a single PDF with those three pages in exactly that order. Ranges work too (3-7), pages may repeat, and - shorthand covers open ends (10- is page 10 to the last page). The plan shows the resulting page list before you run it.

Why are the parts not much smaller than the original?

Because splitting is not compression. Fonts, images and other resources shared across the document are copied into every part that uses them, so the parts together can weigh more than the file they came from. If you need smaller files, that is a separate step — this page only cuts.

Can I see the pages before I choose them?

Not yet. This first version has no thumbnails: it shows the page count, and you choose by page number. Open the file in your PDF reader to confirm which page is which, then type the numbers here. The plan lists exactly which pages land in each part before anything is written.

Do bookmarks, links and form fields survive? Does a signature?

Bookmarks, links to other pages, form fields and accessibility tags live at the document level rather than on the pages, so they may not carry over into the parts — check the result if you rely on them. A digital signature is different: splitting changes the bytes it signed, so it will no longer verify, and that is true of any splitting tool.