How it works
PDF Page Toolkit lets you merge PDFs, split them, and reorder, rotate or delete individual pages using a visual grid of thumbnails. It is free, has no signup, adds no watermark, and does not limit how many files you process.
Your files stay on your computer
This page is a static website. When you drop a PDF onto it, the file is read by
JavaScript running inside your own browser tab, using the standard
File API. There is no server-side component, no account, and no storage.
The PDF is held in your computer's memory for as long as the tab is open, and it is gone
the moment you close or reload the page.
Most other online PDF tools work the opposite way: they upload your document to their servers, process it there, and send a result back. That is a reasonable engineering choice, but it means your document sits on someone else's machine. If you are handling medical records, contracts, payslips or anything else you would not email to a stranger, that difference matters.
How to verify that yourself
Do not take our word for it. Check it:
- Press F12 to open your browser's developer tools.
- Go to the Network tab and click the clear button.
- Now load a PDF, reorder some pages, and save the result.
- Every row that appears is either a file belonging to this site itself, or a
blob:entry. Nothing is addressed to another domain, and nothing carries your document.
Some rows do legitimately appear, and it is worth knowing what they are. To draw the
thumbnails, the browser downloads the page-rendering code
(pdf.worker.mjs), and then, only if your particular document needs them,
supporting files from /pdfjs/ - fonts for PDFs that do not embed their own,
character maps for Chinese, Japanese and Korean text, and decoders for the compressed
images inside scanned documents. Those are downloads of our code, not uploads of your
file. The blob: rows are your browser handing data to itself out of its own
memory: the thumbnails, and your finished download.
Here is the blunt test. Look at the Method column, and switch it on if
your browser hides it by default. Every single request is a GET - a request
to receive something. There is not one POST or PUT in the whole
session, and those are what a page would have to use to send your document anywhere. Check
the Domain column too: every row is this site. Nothing is addressed
anywhere else, ever.
Why the output keeps its quality
A PDF is a structured document, not a picture. Pages are copied from the source file into the new file at the object level, using the pdf-lib library. Text, fonts, vector graphics and images are carried across untouched.
This is why text in the saved file is still selectable and searchable, and why images keep their original resolution: nothing is re-compressed. Tools that convert each page to an image lose all of that. (The total file size will still change, because you are keeping a different set of pages than you started with.) The only place we draw pictures of pages is the on-screen thumbnail grid, rendered with pdf.js, and those thumbnails never touch the file you save.
What it can do
- Merge - load several PDFs and save the whole grid as one file.
- Reorder - drag any page anywhere, including between documents.
- Rotate - one page, a selection, or everything, 90 degrees at a time.
- Delete - hide pages from the output. Nothing is destructive until you save, and deleted pages can be restored.
- Extract - save just the pages you selected.
- Split - every page as its own PDF, one PDF per source file, or a single split point. Multiple files arrive as a ZIP.
Honest limits
- Password-protected PDFs cannot be opened. The tool will tell you clearly when a file is encrypted. It will not attempt to break the password. Remove the protection in the application that created the file first.
- Very large files are limited by your browser's memory, not by an upload cap. Each file needs roughly twice its own size in memory while it is open, so big documents add up. The tool warns you above 200 MB, and reports a readable error if loading fails - but if the browser genuinely runs out of memory, the tab can become unresponsive, and no web page can catch that.
- Rendering thumbnails for very long documents takes CPU. Thumbnails are drawn only as you scroll to them, so the tool stays usable, but a 1,000-page file will keep your fan busy if you scroll through all of it.
- This tool edits pages, not content. It does not edit text, fill forms, sign, compress or convert to Word.
Nothing is tracked to you
See the privacy policy for exactly what is and is not collected.