Merge, split, and reorder PDFs without uploading them
How local browser PDF editing works, when to split or extract pages, and what document features are not preserved.
Most online PDF tools work by uploading your document to a server, processing it there, and sending it back. That is convenient, but it means your contract, medical letter, payslip, or ID scan passes through someone else’s infrastructure, and you rely on their promise to delete it. The PDF tools on this site never do that: the document is opened, changed, and saved inside your browser. This guide explains how that works, which operation to use for which job, and what does not survive the process.
What “local processing” means
The PDF engine is JavaScript that runs on your device. Your file is read from disk into browser memory, the new document is built there, and the download is produced locally. There is no upload step, no account, and no copy on a server. You can confirm this in your browser’s developer tools: the Network panel shows no request carrying your file.
The practical limits follow from that design. Each operation accepts one unencrypted PDF of up to 50 MiB and 500 pages, or, for merge PDFs, up to 20 files and 50 MiB combined. That covers almost all everyday documents while keeping the browser responsive.
Pick the right operation
| You want to… | Use |
|---|---|
| Combine several PDFs into one | Merge PDFs |
| Send only some pages, as one file | Extract PDF pages |
| Turn every page into its own file | Split PDF |
| Drop a few pages and keep the rest | Delete PDF pages |
| Fix the page order | Reorder PDF pages |
| Fix sideways pages | Rotate PDF pages |
| Number pages or mark a draft | Add page numbers, Watermark PDF |
| Make a PDF from photos or scans | Images to PDF |
Page ranges
Pages are selected with the familiar print-dialog syntax: 1, 3-5, 8. Numbers refer to page positions as your PDF viewer counts them, starting at 1. They are not the numbers printed on the pages, which often differ in documents with a cover or front matter. Where order matters, the tools follow the order you type: extracting 3, 1-2 puts page 3 first. Duplicate pages and descending ranges like 5-3 are rejected rather than guessed, so a typo cannot silently produce a different document.
Worked example: an application pack
- Merge the cover letter (2 pages), CV (3 pages), and certificate (1 page) into one 6-page PDF.
- Realise the certificate should come after the letter: reorder with
1-2, 6, 3-5. - Number pages 3 to 6 starting at 1 so the CV and certificate are numbered but the letter is not.
- Open the result and check every page before sending.
What survives, and what does not
Pages are copied, not re-rendered. Text stays selectable and searchable, fonts stay embedded, and vector graphics stay sharp at any zoom. File sizes stay close to the originals, and nothing is converted to images.
Some document-level features belong to the file rather than to pages, and they are not carried over when pages are copied into a new document: bookmarks and outlines, attachments, page labels, document metadata, and form fields. Internal links to pages that were removed stop working. Rotating, numbering, and watermarking rewrite the existing file instead, but any change to a PDF invalidates digital signatures.
Encrypted PDFs and interactive forms are not supported. If a document will not open, remove the password in the application that created it, or print it to a new PDF (Print → Save as PDF), which also flattens forms.
A sensible order of operations
- Merge or split first, so later steps apply to the right pages.
- Delete and reorder pages, then check the page count.
- Apply rotation, page numbers, and watermarks last, since they are easiest to judge on the final page order.
- Sign the document, if needed, only after all edits.
Deleting is not redacting
Removing pages removes them entirely. But drawing a black box over text on a page you keep does not remove the text underneath. It can still be selected, copied, and read by search engines, and several high-profile document leaks happened exactly this way. To hide information on a page, use a proper redaction feature in a PDF editor, which deletes the underlying content, then check by trying to select the redacted area.