Split PDF

Choose one PDF and either export every page as its own file or type ranges like 1-2, 3-5. Each range becomes a separate PDF, and the whole split runs on your own device.

Drop files here or choose files
Files aren't uploaded · processed in your browser · one PDF up to 50 MB
How to split

Every page is exported as its own PDF, and your original file isn't changed.

How to use

  1. Choose your PDF: Drop one PDF onto the box or click to pick it. One file at a time, up to 50 MB. Once it is selected, the page count appears next to the file name, so you don't need to know it beforehand.
  2. Pick how to split: Every page as a file is the default and needs no typing: a 10-page PDF becomes 10 single-page PDFs. Choose Custom ranges instead to type page ranges such as 1-2, 3-5 — each range produces its own file.
  3. Split and save the parts: Click Split PDF. Use the numbered buttons to preview each part, then download it. Downloaded files are named split-1.pdf, split-2.pdf and so on. With a single part there are no numbered buttons — the preview appears straight away. Download saves the part you are previewing; Download all saves them one after another.

How this PDF splitter works

Split PDF takes one PDF and writes out several new PDFs, one for each page or page range you ask for. The splitting code is downloaded with the page and runs inside your browser using the pdf-lib library, so the file is read straight from your disk into memory and the finished parts are built on your own computer or phone. There is no upload step and no server copy, which is why the tool is usable for contracts, payslips, medical letters and ID scans.

Your original file is never modified. The source document is only read — the tool asks it for a page count and copies the pages you selected into brand-new, empty PDFs. Whatever happens, the PDF sitting in your Downloads or Documents folder stays exactly as it was, so there is nothing to undo if you get a range wrong.

Pages are copied whole rather than re-rendered. Text stays real text that you can select, copy and search, images keep their original resolution, page size and orientation are unchanged, and a page that was rotated 90° arrives rotated 90°. Nothing is turned into a picture and nothing is re-compressed, which is the main difference between splitting a PDF and printing selected pages to a new PDF.

The work happens on the page's main thread, so a long or image-heavy document can make the tab feel unresponsive for a moment while it runs. The button shows Splitting… and is disabled until the parts are ready; that is normal, and nothing is lost if you simply wait.

Every page as a file, or custom ranges?

Every page as a file → N pages in, N single-page PDFs out
Custom ranges  1-2, 3-5 → 2 PDFs: pages 1-2, and pages 3-5
Custom ranges  4        → 1 PDF containing only page 4

Pick Every page as a file when you want each sheet on its own: separating a batch of scanned receipts, feeding single pages into an upload form that accepts one document per field, or breaking a signed bundle back into the individual documents it was made from. This mode counts the pages for you at the moment you click, so it works on a file whose length you have never checked.

Pick Custom ranges when you only want part of the document — the two pages of a contract that need signing, a single certificate buried in a 60-page application, or chapters that belong to different people. Type the ranges separated by commas, and remember that each range becomes a separate output file rather than one combined document. If you want pages 1-3 and page 5 inside a single PDF, split them out here and then join the two parts with the Merge PDF tool.

One practical trap: the range box counts the PDF's own pages, not the numbers printed on the paper. Cover sheets, blank inserts and roman-numbered front matter push the printed numbers out of step, so page "1" of a report can easily be the PDF's page 5. Check in your PDF viewer which page number the viewer itself shows before typing a range.

What the Page ranges box accepts

A range is either two page numbers joined by a hyphen (3-7) or a single page number on its own (4). Separate several ranges with commas, and spaces around the numbers are fine: 1-2, 3-5 and 1 - 2,3 - 5 are read the same way. Ranges include both ends, so 1-3 gives you pages 1, 2 and 3. Anything else — letters, 1–3 with a dash copied from a word processor, or an open-ended 5- — is rejected with the message "Check the range format, for example: 1-2, 3-5".

Page numbering starts at 1 and ranges always run from the lower page to the higher one, so a 0, a page number larger than the document's length, and a backwards range such as 3-1 all come back as "Page numbers must be between 1 and …" with your document's real page count filled in — which is also the quickest way to confirm how long the PDF actually is. Decimals such as 1.5 are not page numbers at all and fail the format check instead.

An invalid range stops the whole job rather than returning the parts that would have worked. That is deliberate: you either get the complete set of files you asked for, or an error message and nothing else, so you never download a half-finished set without noticing. Fix the one bad number and click Split PDF again.

Ranges may overlap and may repeat, and they are processed strictly in the order you typed them. 1-3, 2-4, 1-3 produces three files, the first and third being identical copies of pages 1-3. 2-3, 1-1 produces a first file holding pages 2-3 and a second file holding page 1 — the tool does not sort your ranges for you, so the numbering of split-1.pdf, split-2.pdf and so on follows your list, not the page order.

What the split files keep — and what they lose

Everything on the selected pages carries over: text and fonts, vector drawings, images at full resolution, page dimensions, orientation and rotation, and the original order of the pages within each range. Because the pages are copied as they are, a split part looks identical to those pages in the original document at any zoom level.

Document-level extras do not carry over, because they belong to the original file rather than to any individual page. Bookmarks and the clickable outline in a viewer's sidebar are gone, file attachments are not copied, the document-level structure behind fillable forms is not rebuilt, and the title and author shown in the file properties come out empty — the parts report pdf-lib as the program that produced them. Internal links also break when their target is not in the same part: a table of contents that jumps to page 40 has nothing to jump to in a file that only contains pages 1-5.

Splitting does not shrink a file in proportion to its pages, which surprises people who split a PDF hoping to get under an upload limit. Resources shared across pages — embedded images and fonts — are copied into every output that needs them. In our test, a 4-page image-heavy PDF of 825 KB produced four single-page files of about 823 KB each, roughly four times the original in total. A 5-page text-only PDF of 4,052 bytes behaved as expected and produced five files of about 1,086 bytes. If size is the real problem, run the parts through the Compress PDF tool afterwards.

Password-protected PDFs are refused before anything is split, with the message "This PDF is password-protected. Remove the password first, then try again." This tool never asks for a password and cannot decrypt a file. The output files are unencrypted. Encrypted source PDFs, including those with permission restrictions, must first be saved as an unprotected copy: open the document with its password in a PDF reader, save or print it as a new unprotected PDF, and split that copy instead.

Saving the parts, and what this tool can't do

The parts are kept only in your browser's memory and are handed to you as local download links. Download saves one part; Download all saves them in sequence, and because those are separate downloads your browser may ask whether to allow multiple files from this site — say yes, or save the parts one at a time. There is no ZIP archive: the files arrive individually, named split-1.pdf, split-2.pdf and so on, and renaming is done afterwards in your file manager. Nothing is written to storage or cookies, so refreshing or closing the tab clears the file and all the parts, and Start over clears them immediately.

A few things are outside what this tool does. It splits one PDF at a time, so a folder of documents means one pass each. It splits by page number only — not by file size, by bookmark, or automatically every N pages. It shows no page thumbnails to click, so selecting pages means typing their numbers or using Every page as a file. It cannot merge chosen ranges into one document, reorder pages, delete pages in place, or convert the output to images or Word: every part it produces is a PDF.

Frequently asked questions

Is my PDF uploaded when I split it?
No. The file is read and split inside your browser by code that was downloaded with the page, and no copy is sent to a server. The parts live in the tab's memory until you download them, and closing or refreshing the tab clears everything.
Can I extract pages 1-3 and 5 into a single PDF?
Not in one step. Each range you type becomes its own file, so 1-3, 5 gives you two PDFs. Split them here, then combine the two parts with the Merge PDF tool to get one document.
Why are my split files almost as big as the original?
Images and fonts shared by several pages are copied into every part that uses them. In our test a 4-page image-heavy PDF of 825 KB produced four files of about 823 KB each. In our 5-page text-only sample, each output was about 1,086–1,091 bytes, compared with 4,052 bytes for the original. Other PDFs may behave differently; check each output's size, and compress the parts afterwards if they are still too large.
Can I download all the parts as one ZIP file?
No, there is no ZIP option. Download all triggers the downloads one after another, so your browser may ask permission to save multiple files. You can also download each part separately from the result list.
What does "Page numbers must be between 1 and …" mean?
One of your ranges points outside the document — a page number higher than the PDF's length, a 0, or a backwards range such as 3-1. Pages are numbered from 1 and ranges run low to high. Correct the number and click Split PDF again; nothing is produced until every range is valid.
Do bookmarks, form fields and the document title survive the split?
No. Those belong to the whole document, so the new files start without bookmarks, attachments or the original title and author, and fillable forms may stop working. Page content, text, images, page size and rotation are all kept exactly as they were.
Can I split a password-protected PDF?
No. Encrypted PDFs are rejected before splitting and the tool never asks for a password. Open the file with its password in a PDF reader, save it as a new PDF without protection, and split that copy.