Split PDF takes one document and gives you back only the pages you asked for. Type a range like 1-3, 5 and choose whether you want a separate file for each range or a single PDF containing all of them. Everything happens inside your browser: the file is read into memory, rebuilt there and handed straight back to you as a download, so nothing is ever uploaded to a server. There is no watermark, no page cap and no signup, and the pages are copied rather than re-rendered, so text stays selectable and images keep their original quality.
How to split a PDF
- 1 Add the PDF you want to split — the page count appears once it is read.
- 2 Type the pages to extract, for example 1-3, 5, 9- for pages 1 to 3, page 5 and page 9 to the end.
- 3 Choose one file per range, or a single PDF containing every selected page.
- 4 Click Split and download the result — several files arrive together as a .zip.
How splitting works in your browser
When you add a file, it is read into memory with the File API and parsed by pdf-lib, a PDF library written entirely in JavaScript. Reading it is what produces the page count shown next to the file name — nothing is sent anywhere to obtain it. The library is loaded on demand, the moment you drop a document, rather than shipped with the page, which is why the tool opens instantly even though the engine behind it is not small.
The range you type is parsed into a list of page indices. For each output document, a new empty PDF is created and the selected pages are copied into it one by one. Copying a page brings its content stream along with the fonts, images and colour spaces it references, which is why the result is a genuine PDF rather than a picture of one. Between pages the tool hands control back to the browser for a frame, which is what keeps the progress bar moving and the tab responsive on a document with hundreds of pages.
Each finished document is serialised back to bytes and offered as a download through a temporary blob URL, which is revoked immediately afterwards so the memory is released. When there is more than one output they are packed into a .zip first, again locally, so a five-way split is still a single click.
What people extract
One invoice from a statement
Annual statements arrive as a single file and accounting wants the one month. Extracting page seven takes a moment and avoids emailing a full year of transactions to prove a single charge.
The pages of a contract that concern one party
Send the schedule and the signature page without the eighty pages of boilerplate. The recipient gets a document they can actually read, and nothing they were not meant to see.
Chapters from a scanned book or manual
A 400-page scanned manual becomes one file per section, each small enough to attach to an email and each named after the range it came from.
Separating a batch scan
A feeder scanner produces one file containing twenty unrelated documents. Splitting on the boundaries turns it back into twenty files, without the round trip through an upload service.
Why split a PDF in the browser
Splitting is usually the last step before sending something to someone else: the two pages of a contract that actually concern them, the invoice for one month out of a twelve-month statement, the chapter a colleague asked for. Those are precisely the documents that should not be passing through a stranger's server on the way. Because KitYards does the work locally, the file never leaves your machine, and you get a clean result with no watermark and nothing to sign up for.
What it cannot do
Running entirely in a browser is what makes this private, and it is also where the limits come from:
- Password-protected PDFs are refused rather than mangled. pdf-lib cannot decrypt content streams, so a document that needs a password to open would come out blank or garbled; the tool says so instead. Remove the password in your PDF reader first.
- Very large files are bounded by your device's memory. The whole document is held in RAM while it is being split, so a several-hundred-megabyte scan can exhaust a tab where a desktop application would stream through it.
- It splits by page number, not by content. There is no detection of chapter breaks, blank separator sheets or barcode markers — you tell it where the boundaries are.
- Interactive form fields and digital signatures do not survive. Copied pages carry their appearance but not the form structure, and any cryptographic signature is invalidated, which is correct: the document it signed no longer exists.
- Bookmarks, the table of contents and document-level metadata are not carried into the output. Each result is a clean document with KitYards as its producer.
Troubleshooting
- "Those pages are outside the document"
- The range asks for a page the file does not have. Check the page count shown next to the file name — it is the real number, not the one printed in the footer of the document, which often starts counting after a cover and a table of contents.
- The file will not open at all
- Almost always password protection or corruption. Open it in your normal PDF reader: if it asks for a password, remove it there and add the file again. If the reader also refuses it, the file is damaged and no splitter can recover it.
- The tab slowed to a crawl on a big document
- Memory pressure. Extract fewer ranges at a time and download between them, and close other heavy tabs — they share the same memory budget. A document of a few hundred pages is fine; one of several thousand may not be.
Frequently asked questions
How do I write the page range?
Use numbers and hyphens, separated by commas. 1-3 means pages one to three, 5 means just page five, and 9- means page nine to the end of the document. You can combine them: 1-3, 5, 9- is all three at once.
What is the difference between the two output options?
"One file per range" gives you a separate PDF for each comma-separated part, which is what most people mean by splitting; with more than one range they arrive together in a .zip. "A single PDF" puts every page you selected into one new document instead, which is really an extract.
Is my document uploaded anywhere?
No. You can confirm it in about ten seconds: open your browser's developer tools, switch to the Network tab and split a file. No request carries your data. You can also disconnect from the internet after the page loads — the split still completes normally.
Does splitting reduce quality?
No. Pages are copied as objects rather than re-rendered, so text remains real text, vector graphics stay sharp and embedded images keep their original compression. Nothing is rasterised along the way.
Why is my one-page extract not one tenth the size of the original?
Because a PDF shares resources across pages. A single page that uses three embedded fonts carries those fonts with it, even if the other forty pages used them too. The page content shrinks; the shared resources it depends on largely do not.