Free · in your browser · no signup
Split PDF — Extract Pages Free
Pull out specific pages or split a PDF into separate files — privately, in your browser. Nothing is uploaded.
⬆
Drop PDF files here
Drop one PDF to split or extract pages from
We use cookies and similar technologies to enhance your browsing experience, analyze site traffic, and personalize content and ads. By clicking "Accept", you consent to our use of cookies. Learn more in our Privacy Policy.
Pull out specific pages or split a PDF into separate files — privately, in your browser. Nothing is uploaded.
Split PDF takes a single PDF and breaks it into smaller files. You can pull out a specific page range (say pages 12 through 18 of a contract), grab a handful of non-contiguous pages, or burst every page into its own one-page file. It's a common chore when you only need to share part of a scanned report, separate the chapters of a bundled export, or strip a cover sheet off a document before sending it on.
Everything happens in your browser. When you choose a file, it's read directly into the page's memory and processed on your machine; the bytes are never sent to a server. That matters because the PDFs people split are often things they'd rather not upload anywhere: signed agreements, payslips, medical forms, internal drafts. You get the same result as a desktop tool without installing anything or trusting an upload.
Pick a PDF, tell the tool which pages you want, and download the result. There are two modes:
document-1.pdf, document-2.pdf, and so on), usually delivered as a ZIP so you don't fire off dozens of downloads.Under the hood the original file is parsed into its page objects, the pages you asked for are copied into a fresh document, and that document is re-serialized to bytes you can save. The source file is never modified. Because the work runs on your device, file size is bounded by your available memory rather than an upload limit, so large scans are fine as long as your browser tab has room for them.
Say you have report.pdf with 30 pages and you only need the executive summary on pages 3 to 5 plus the appendix on page 28.
Enter a page expression like:
3-5, 28
The tool reads that as pages 3, 4, 5, and 28, copies those four pages into a new document, and gives you a 4-page PDF. Page numbers are 1-based (the first page is 1, matching what a PDF viewer shows you), and order follows the original document, so listing 28, 3-5 produces the same page order as 3-5, 28. If you'd rather have each of those four pages as its own file, switch to burst mode and you'll get four separate one-page PDFs instead. An out-of-range number (like 45 on a 30-page file) is flagged rather than silently dropped, so you can fix the range before exporting.
Because nothing leaves your device, it's also a sensible default for documents with personal or confidential content, where uploading to an unknown web service isn't acceptable.
Splitting copies pages, it doesn't shrink them. A page that's a 4 MB scanned image is still 4 MB in the output. If your split file is larger than expected, the page content is the cause, not the split itself.
Encrypted or password-protected PDFs won't open until the password is supplied or removed. A document can be readable in your viewer but still carry restrictions that block page extraction.
Form fields, annotations, and digital signatures may not survive cleanly. Splitting rebuilds the document from its pages; interactive form data and signature validity can be lost in that process, which is expected behavior for any tool that re-serializes a PDF.
Page numbers are physical positions, not printed labels. If a document's printed footer says "Page 1" on the third sheet (common with cover and TOC pages), use the position the viewer shows, not the printed number.
A PDF isn't a stream of pages the way a text file is a stream of lines. It's a collection of numbered objects (fonts, images, content streams, page dictionaries) tied together by a page tree and an index called the cross-reference table. Each page is a dictionary that points at the resources it needs.
Splitting works by walking that tree, picking the page dictionaries you asked for, and copying them plus everything they reference into a new object set with its own page tree and cross-reference table. Shared resources like an embedded font are copied as needed so each output is self-contained.
This is why splitting is fast and lossless for the page content itself: the bytes of each page's content stream are carried over verbatim rather than re-rendered. It's also why metadata that lives at the document level (the outline/bookmarks, document-wide form definitions) doesn't always carry across, since it isn't owned by any single page.
Choose extract mode and enter just that page number, for example `7`. The result is a one-page PDF containing only that page.
Yes. Burst mode creates one file per page, typically packaged as a ZIP so you get all of them in a single download.
No. The PDF is read and processed entirely in your browser on your own device. Nothing is sent to a server.
There's no fixed cap. The practical limit is your device's available memory, since the whole file is held and processed in the browser tab.
Encrypted PDFs block page extraction until the password is provided or removed. Open it in a viewer with the password and save an unprotected copy first.
Page content carries over reliably, but document-level data like bookmarks and interactive form definitions, and digital signature validity, may be lost when the file is rebuilt from its pages.
Only by dropping the pages you didn't keep. The pages you extract retain their original size, so a split of image-heavy pages stays large.
Use the physical position shown by your PDF viewer (1-based). Printed footer numbers can differ when a document has unnumbered cover or contents pages.