Split by Size

Divide a long PDF into numbered parts, each one measured against the limit you set.

Hybrid processingPDFWorkingNo account needed

What this tool does

Split by Size fills a new document one page at a time and saves it after every page to see how large it has actually become. When the next page would carry the part past your limit, that page is taken back out, the part is written and closed, and a fresh one opens with that page as its first. Because each candidate is serialised and weighed for real rather than estimated from a page count or from the source file's bytes, the size shown on a part is the size it occupies on disk.

The limit is one number and one unit. MB and KB are read as 1024-based, so 5 MB means 5,242,880 bytes and 500 KB means 512,000. Anything under 1 KB is refused outright, which is worth knowing if you switch the unit to KB: the field steps in 0.05, and 0.05 KB comes to about 51 bytes. Parts come out in document order, named after the source with -part1, -part2 and so on, and every page lands in exactly one of them — nothing is repeated across the set and nothing is left behind. When more than one part is produced they can be saved singly or taken down together as a ZIP.

Two things decide whether this is the right tool for the file in front of you. A single page larger than your figure on its own cannot be divided further; it is written as its own part, still over the limit, and the run summary counts how many pages that happened to. And splitting removes no weight: each part re-embeds the fonts and images its own pages call on, so a typeface used throughout the document is written into every part that uses it, and the parts added together will usually come to more than the file you started with. When one file has to meet a figure, Compress PDF to Target Size is the tool that takes weight out rather than dividing it up.

The whole job runs in this browser tab through pdf-lib. The document is read into memory here, each part is built and measured here, and no copy is sent anywhere — which matters when the thing being cut up is a contract, a case bundle or a medical record. What it costs is the document-level furniture: every part is a newly created file holding copied pages, so the outline, the title and author fields and a form's own dictionary are not carried across, and an internal link aimed at a page that landed in a different part has nothing left to resolve to. Speed is the other trade, since the part being built is re-saved every time a page joins it — and once more for any page that has to start a part of its own — so several hundred pages takes noticeably longer than a plain extraction.

How to use it

  1. Choose the PDF to divide. One document per job — add several and each is split in turn into its own numbered set. A file that asks for a password cannot be read, so remove it with Unlock PDF first.
  2. Set the maximum size per part. Type the figure, then choose MB or KB in the Unit menu — the chip printed beside the number box always reads MB, so it is the menu that decides the unit, not the chip. The default is 5 MB; mailbox ceilings usually sit between 10 and 25 MB, and leaving headroom is sensible because encoding inflates an attachment in transit.
  3. Run it and read what it reports. The line above the parts gives the pages that went in, the number of parts that came out, and how many pages were larger than the limit on their own.
  4. Save the parts you need. Each part downloads on its own, or the whole set comes down as one ZIP when the split produced more than a single file.

Processing & privacy

Hybrid processing. Smaller jobs can run locally; heavier jobs may use secure server processing.

Common use cases

Getting a long report past an email limit

A 40 MB document will not leave most mail servers. Split at 8 MB and it becomes five or six attachments that will, still in reading order.

Uploading to a portal that caps each file

Filing systems, grant portals and tender submissions often refuse anything over a fixed size. Numbered parts satisfy the cap without you working out page ranges by hand.

Handing over a scanned archive in instalments

A 600-page scan is an unwieldy single download. Fixed-size parts are something a colleague on a poor connection can fetch one at a time.

Keeping an attachment set predictable

When a recipient's system is known to reject large files, splitting to a figure below their ceiling avoids the bounce that arrives an hour later.

Questions

Is the PDF uploaded in order to be split?
No. The split runs in this browser tab through pdf-lib — the file is read into memory here, the parts are assembled and measured here, and there is no server operation behind this tool for a document to be sent to. Splitting at each top-level bookmark is the one that genuinely needs the processing service, and its own page says so.
Can I chunk PDF files to fit an email attachment limit?
That is the common reason to chunk PDF documents at all, and it works. Set the limit somewhat below what the mail server accepts, because attachments are base64-encoded in transit and grow by roughly a third on the way.
Why is one of my parts still over the limit?
Because a single page was bigger than your figure on its own. Pages are never divided, so that page is written as its own part and counted in the summary. Compressing the document first is the way to bring such a page down.
Why do the parts add up to more than the original?
Each part is a self-contained PDF that re-embeds the fonts and images its own pages use, so anything shared across the document gets written more than once. This tool divides a file; it does not compress it.
Do bookmarks and form fields survive the split?
No. Every part is created fresh from copied pages, so the outline, the document's title and author fields and the dictionary that makes form fields work are not carried over. Page content and the annotations sitting on each page are.
Can I split at chapters or at a phrase instead?
Not here — this one counts bytes and nothing else. Split by Bookmark cuts at each top-level bookmark, Split by Text cuts wherever a page matches a phrase, and Extract Pages handles a range you can name yourself.
How large a document can it cope with?
No ceiling is imposed, so memory and patience are the real constraints. Every page that joins a part forces that part to be written out again so it can be weighed, and that work grows with the length of the document.

Related tools

  • Merge PDF — Combine several PDFs into one document in the order you choose.
  • Split PDF — Split one PDF into separate documents by page range.
  • Compress PDF — Reduce PDF file size while preserving the quality that matters.
  • PDF to JPG — Export every page of a PDF as a JPG image.
  • JPG to PDF — Turn one or many JPG photos into a single PDF.
  • Protect PDF — Add a password and set permissions on a PDF.