PDFWren

Split a PDF into parts under a size limit

Break a large PDF into parts that each fit under a size limit. Every part is built and measured in your browser, so the sizes shown are the sizes you get.

🔒 Runs in your browser

When a file is too big and compressing is not the answer

Sometimes you cannot compress. A court wants the exhibits legible. A regulator specifies a scan resolution. The document is already as small as it can usefully go and it is still over the cap.

Then the answer is volumes: split the document into parts that each fit, and submit them in sequence. That is standard practice in legal filing, and it is the fallback almost every upload form expects.

Where the limits come from

They vary far more than people assume, and there is rarely a universal number to memorise.

US federal courts are the clearest illustration. Each district runs its own CM/ECF instance and sets its own maximum PDF size, and the range across districts is wide — some courts cap a single document in the single-digit megabytes while others allow a hundred. Several have raised their limits in recent years, and announced it as news.

The practical consequence is simple: check your own court's current limit on its own website, not a general article, and not this one. The same applies to government portals and email providers.

What this tool does is take whatever number you were given and guarantee the parts respect it.

Measured, not estimated

This is the part that matters, and it is where most tools quietly cut a corner.

The easy implementation divides total file size by page count and packs that many pages per part. It is wrong. PDFs routinely share resources between pages, so a two-page extract from a four-page scan can come out nearly the size of the original — we measured exactly that on a real document: 2 pages pulled from a 2,205 KB scan produced a 2,203 KB file, because all four page images came along.

An average has no relationship to what a part actually weighs. So every part here is genuinely assembled and weighed, and only then compared against your limit. The sizes you are shown are the bytes you download.

That matters most in exactly the situation you are probably in: a deadline, an upload form, and no appetite for finding out at submission time that "24 MB" was actually 31.

Compress first, then split

If the parts come back numerous or barely smaller than the original, the document is image-heavy and compression will do more than splitting can.

The sensible order is to compress to a target size first, then split what remains if it is still over the cap. Doing it the other way round means compressing several parts individually and hoping each lands.

Practical notes

  • Name the parts clearly when you submit them. Most filing systems expect something like "Volume 1 of 4" in the description, and reviewers need to know the order.
  • Check whether splitting is allowed at all. Some systems require a separate caption or cover page on each volume, and some ask you to call the clerk first. Splitting a document that should have been compressed can get a filing rejected just as easily.
  • Your original is untouched. Every part is a new file; nothing overwrites what is on your disk.

This tool runs entirely inside your browser. Your file is never uploaded — it never leaves your device, so there is nothing for us or anyone else to read, store or leak.

Common questions

How is this different from splitting by page count?

Splitting by pages gives you equal numbers of pages, which tells you nothing about file size — a scanned photograph page can be fifty times heavier than a page of text. This splits by what the upload form actually cares about: bytes. Each part is built and weighed before you receive it.

Are the parts really under the limit?

Yes, and that is the whole point of the design. Every part is fully assembled and serialised on your device, and the real byte length is what gets tested against your limit. Nothing here is estimated from a per-page average, because per-page averages are badly wrong on the documents that need splitting most.

What if a single page is bigger than my limit?

A page cannot be divided, so it is emitted as its own part and you are told which pages are over. The fix is to compress the document first and then split it — a high-resolution scan of one page can easily exceed a 10 MB cap on its own.

Why did I get parts almost as big as the original?

Some PDFs reference all their images from one shared dictionary, so a part carries images it never displays. The split is still correct, but it has bought you little. The tool detects this and says so. Compressing before splitting gives you far fewer, smaller parts.

Does splitting reduce quality?

No. Pages are copied across as they are, with no rendering and no re-encoding. A split scan is pixel-identical to the original pages. Splitting only removes pages from a part; it never degrades the ones it keeps.

Written and maintained by Mustak Momin · Last reviewed