Extract pages from a PDF into a new file
Extracting pages from a PDF copies only the pages you select — with their fonts, images and annotations — into a brand-new document, leaving the source file untouched. This tool accepts page numbers and ranges like 2, 5-9, 14 and builds the new PDF entirely in your browser, so nothing is uploaded. It is the fastest way to pull one chapter, exhibit or statement out of a long document.
Extraction and deletion are two ways of describing the same cut. If the pages you want to keep are few, name them here; if the pages you want to lose are few, the remove PDF pages tool is quicker. And when you need every range saved as its own file — say chapters 1, 2 and 3 as three PDFs — that is the job of Split PDF.
How it works: copying page objects across documents
PDF pages are not self-contained blobs; each Page object is the root of a small graph that references a content stream, a resources dictionary, embedded font programs, image XObjects and any annotations sitting on the page. Copying a page correctly means copying that entire graph and rewriting every internal object reference to point at the copies — which is exactly what pdf-lib’s copyPages does under the hood of this tool.
The tool parses your selection string into a sorted list of page indices, creates a fresh empty document, and imports each selected page’s object graph into it. Object numbering starts over in the new file, references are remapped, and an object needed by two selected pages — a shared font subset is the classic case — is imported once and referenced twice, never duplicated. Because objects are copied verbatim rather than re-encoded, the extracted pages are pixel-for-pixel and glyph-for-glyph identical to the originals: text remains selectable, vector art remains sharp at any zoom, and image compression is untouched.
What does not come along: document-level structure that lives outside the selected pages. Bookmarks pointing into unselected pages, document metadata beyond the basics, and named destinations elsewhere in the file are left behind, since carrying them over would create dangling references. The output is a clean, minimal document containing exactly the pages you asked for and the resources they genuinely use.
Worked example: pulling one chapter from a 214-page report
An annual report runs 214 pages at 27.3 MB. You need only chapter 4 — pages 96–123 — plus the summary table on page 209, to send to a colleague. Entering 96-123, 209 extracts 29 pages into a new file of 3.1 MB. The size dropped far more than the page ratio (29/214 would predict about 3.7 MB) because the heaviest photo spreads live in other chapters, while chapter 4 is mostly text over two shared fonts, which are copied exactly once. The extraction completed in roughly two seconds, and the resulting file opens showing page 1 — its own numbering — even though the content began life as page 96.
Extracting PDF pages — frequently asked questions
What is the difference between extracting and removing PDF pages?
Extraction keeps only the pages you name; removal deletes the pages you name and keeps everything else. They are complements: extracting pages 5-9 from a 20-page file equals removing pages 1-4 and 10-20. Pick whichever list is shorter to type.
Do extracted pages keep their original quality and fonts?
Yes. The pages are copied as objects, not re-rendered, so text stays selectable and images keep their original resolution and compression. Embedded font subsets travel with the pages that use them.
Can I extract pages in a different order, like 7 then 2?
This tool writes extracted pages in ascending document order to keep the operation predictable. To change the sequence afterward, run the result through the rearrange pages tool.
Why is my extracted PDF larger than I expected?
Every extracted page brings along the fonts and images it references, so a single page can carry a sizeable font or a full-page graphic. Ten pages that share one 400 KB font still need that font once, no matter how few pages you take.
Is the original PDF changed when I extract pages from it?
No. The source file is only read, and the selected pages are copied into a separate new document that you download. Your original keeps all of its pages.
Can I extract pages from a password-protected PDF?
Not while it is encrypted, because the page objects cannot be read without the password. Open it with the password and save an unlocked copy first, then extract from that copy.