Delete PDF pages
Pick pages in a thumbnail grid and remove the ones you do not want.
or drop files anywhere on this page
Your files stay in this browser. Nothing is uploaded.
About Delete Pages
Delete PDF pages lets you delete pages from a PDF, or from several at once, by marking them on a grid of page thumbnails, all in your browser. Reach for it to drop a blank scanner page, an outdated cover sheet, or a stray page that ended up in the wrong file, without opening the whole document in an editor.
Nothing is removed until you click Apply changes, and every Delete has a matching Undo, so you can mark pages freely and change your mind before committing. Your original file is never touched, and the pages you keep are not re-rendered or recompressed.
How it works
- Add one or more PDFs. Every page appears as a thumbnail, and the Thumbnail size slider makes them larger when you need to read a page before it goes.
- Click Delete under any page you want to remove. The page is veiled and its button turns into Undo.
- For a lot of pages at once, use "Removing a lot of pages? Type a range", which appears when you have added a single file. Write pages and ranges such as "2,4,9-12", or an open range such as "45-" for page 45 to the end.
- Turn on Discard bookmarks if you want to drop the outline entirely.
- Click Apply changes and download your result.
What is removed and what is kept
The result is built from the pages you keep, so a marked page's text is not written into the saved file, and neither are the pictures and fonts only that page used, even in a file whose pages share one list of resources. Pages you keep pass through untouched: text stays text and images stay as they were.
Bookmarks are looked after. One that pointed at a deleted page is dropped, and the rest are repointed to their new page numbers. When a parent bookmark goes, its children move up to take its place. Discard bookmarks removes the outline altogether.
The result keeps your file's name, and several files come back together as one ZIP.
What does not come across
Because the file is rebuilt, a few things are left behind. A link or a bookmark that pointed at a page you deleted is dropped rather than left pointing at nothing, and a link that found its target by name, instead of pointing straight at a page, stops working: the named destinations themselves do not come across. Neither does anything the document was set to do on its own — its JavaScript, the page it opened at — nor the tagged-PDF structure a screen reader takes its reading order and alt text from, nor article threads.
What describes the document as a whole does come across: its title and author, its language, how it opens, its attachments, and its XMP metadata — minus the page thumbnails and page counts that packet held, which described the pages as they were. So do the layers the pages you keep are tagged with — a layer that was hidden is still hidden, and one only a deleted page used is gone with it. Page labels are renumbered to follow the pages that survive, so front matter numbered i, ii, iii, 1, 2 with page 2 deleted comes back i, ii, 1, 2. Bookmarks to the pages you kept, and the form fields on those pages with whatever was typed into them, come along too.
Limits, and what to do when something goes wrong
This tool sets no file size or page count limit of its own. The whole file is read into your browser, so the practical limit is your device's memory.
- A document must keep at least one page. If you mark every page of a file, a note appears and Apply changes stops with a message naming the file. Nothing is produced for that batch until you unmark a page.
- The range box appears only when you add a single file. With several files, use the Delete buttons on the grid.
- The range box marks exactly the pages its text names and keeps up as you type, so "12" ends up marking page 12 alone and emptying the box takes its marks back. Pages you marked with their own Delete button are left alone.
- A page number past the end is refused with a message such as "Page 40 is beyond the last page (30)".
- A password-protected PDF asks for its password before it can be used, and a wrong password stops with a message.
A worked example
A 30-page scan has a blank page 12 and a stray fax cover as pages 28 to 30. Add the file, click Delete under page 12, then paste "28-30" into the range box. Four pages now show Undo. Click Apply changes and you download a 26-page PDF.
Frequently asked questions
Usually, yes. The result is built from the pages you keep, so a deleted page's text, images and fonts are not written into it, and dropping a page that held a large photo takes that photo's weight out of the file. What the pages you keep still use stays, so how much you save depends on what the deleted pages were carrying.
Yes, with a check. The result is built from the pages you keep, so a deleted page's text is not written into it, the pictures and fonts only that page used are left out, and a link or bookmark that pointed at that page is dropped rather than dragging it along. An unusual file can still carry a picture across, so open the result and check it before you share it, and run it through Flatten PDF with Flatten everything if nothing of the page may be recoverable.
No. A PDF needs at least one page. If you mark every page of a file, a note appears and Apply changes stops with a message instead of producing an empty document. Unmark at least one page and apply again.
Yes, as long as you have not clicked Apply changes yet: click Undo on that page to restore it. Nothing is written to the output file until you apply the changes, and your original file is never touched at all.
Yes. Add multiple files and each one is processed separately; you get one result per input, downloaded together as a ZIP, with only that file's marked pages removed. The range box is offered only for a single file, so use the Delete buttons when you have several.
They are dropped automatically, and any child bookmarks move up a level. Bookmarks pointing at pages you kept are repointed to the new page numbers, unless you turn on Discard bookmarks to drop the outline entirely.
"Removing a lot of pages? Type a range" marks a whole set of pages at once, like 2,4,9-12, instead of clicking Delete on each one. It understands single pages, ranges, open ranges such as 45- and the words odd and even, and its marks follow what you type, leaving pages you marked with their own Delete button alone.
They reach the same kind of result from opposite ends: Delete Pages marks the pages to drop, Extract Pages marks the pages to keep. Both build the result from the pages that survive, so what only the dropped pages used is normally left out either way. Extract also renames its results and can take all the odd or even pages in one click.
The file is read and rewritten entirely in your browser and is never uploaded. Your browser does keep a Recent files entry on your own device, with a copy of files under 5 MB, which you can remove from the Recent files page. The text of the pages you delete is not written into the result at all.
Related tools
This tool ran entirely in your browser and nothing was uploaded. Your Recent files list keeps a copy of local files under 5 MB until you clear it. How your files are handled.