Extract Pages From a PDF
Pull a chapter, a section, or a single page out of a document and save it as its own PDF. You list the pages to keep and everything else is left behind. The extracted pages are copied rather than re-rendered, so the new file is identical in quality to the original, and the whole operation happens inside a single pass on the server.
Use this without the search next time. Prathom Workbench puts Prathom's tools in your toolbar.
Add to Chrome — freeExtract PDF pages
Pull specific pages into a brand-new PDF; everything else is left out.
What it does
- Extract any pages into one new document
- Ranges, open-ended ranges, and odd/even keywords
- Lossless — pages are copied, not re-rendered
- Live count of what will be extracted
- No watermark, no account, deleted within 30 minutes
How to use Extract PDF Pages
- 1
Open the source PDF
Drop the file in. The page count appears immediately so you can work out the numbers you need.
- 2
List the pages to keep
Type page numbers and ranges separated by commas. Open-ended ranges like "20-" work, as do the keywords odd, even, first and last.
- 3
Confirm the selection
The panel reports how many pages of how many are selected and normalizes what you typed, so overlapping ranges show as a single tidy list.
- 4
Extract and save
Press Extract pages and save the new document. Your source file is not changed.
How it works
A new empty document is created and the pages you selected are copied into it in the order they appear in the source. Copying is a structural operation: each page object arrives with everything it references, and all internal object numbers are rewritten so they resolve correctly in their new document.
Because nothing is interpreted or re-encoded, the extract is not a reproduction — it is the same pages. Text stays as text, vector artwork stays resolution-independent, and an embedded photograph keeps the exact JPEG data it had.
The selection is a set, not a list. Overlapping ranges merge, duplicates collapse, and the result is always sorted ascending. This is why "1-10, 5-15, 12" becomes "1-15" rather than producing pages in a strange order or twice. It also means you cannot use this tool to reorder pages — extraction preserves document order by design.
Pages outside the document are reported rather than silently ignored. Asking for page 50 of a 40-page file produces a message saying so, because the alternative is an extract quietly missing something you asked for.
When you'd use this
Sharing one section of a long document is the common case: sending a supplier just the pages of a contract that concern them, or a colleague the two pages of a report they actually need.
Uploading to a form with a page limit, where only part of a document is required.
Archiving something specific out of a bulk file — one invoice from a monthly statement, one form from a scanned batch.
Preparing a document for printing, where you want a handful of pages rather than paging through a print dialog looking for the right ones.
The size surprise, explained once
People are often startled that extracting one page from a 200-page report gives a file half the size of the original rather than a two-hundredth.
The reason is font embedding. A PDF stores each embedded font once and every page that uses it points at that single copy. Extract one page and the font must come with it, in full, because there is no way to know which glyphs a reader will need. Two or three embedded font families can easily account for most of a small document's size.
Scanned documents behave the way intuition expects, because each page is a self-contained image with nothing shared between them. If your extract seems disproportionately large, an embedded font is almost always why.
Examples
Taking one chapter out of a manual
The output contains only those eighteen pages, renumbered from one inside the new document. Page 31 of the manual becomes page 1 of the extract, which is worth knowing if you are going to refer to page numbers in an email about it.
Overlapping ranges typed by mistake
Three overlapping instructions collapse into a single clean range. No page appears twice, because the selection is a set rather than a list. Seeing "1-15" echoed back is often the moment someone realises they typed the second range wrong.
Frequently asked questions
How is this different from splitting a PDF?
Extraction produces one new document containing the pages you chose. Splitting produces several documents from one. If you want pages 5 to 10 as a single file, extract. If you want each page as its own file, or one file per chapter, split. The two tools sit either side of the same operation.
Are the extracted pages lower quality?
No. The pages are copied as objects, complete with their fonts, vector graphics and embedded images, and nothing is rendered or re-compressed. The extract is visually identical to those pages in the original at any zoom level, and text remains selectable text rather than becoming an image.
Why is my one-page extract nearly as large as the whole document?
Because a page brings its dependencies. If the page uses an embedded font, the whole font has to be copied into the new file even though only a few characters of it are used, and the original was sharing that one copy across every page. This affects text-heavy documents most. Scanned documents, where each page is a self-contained image, extract almost exactly in proportion.
Do the page numbers restart in the new file?
The document's physical page numbering does, so the first page you extracted becomes page 1. Any page numbers printed on the pages themselves are part of the page content and stay exactly as they were — an extract of pages 31 to 48 will show "31" printed on its first page. That mismatch is normal and cannot be avoided without editing the page content.
Can I extract from a password-protected PDF?
No. Encrypted documents cannot be read without the password, and the tool refuses them rather than producing corrupted output. Open the file in a PDF reader with the password, save an unprotected copy, and extract from that.