Skip to content
user@convertitive:~$convertitive

Extract Specific Pages from a PDF

Type a page list — 1, 3, 5-8, 12 — in whatever order you want them. One new PDF comes out.

This is not the same job as splitting a range. Range extraction pulls one contiguous run of pages — “from page 3 to page 7” — in their original order. This tool takes an arbitrary, possibly non-contiguous set of page numbers and ranges, typed in a single pass, and assembles them into one PDF in exactly the order you typed them. Because the order you type is the order you get, you can also reorder pages relative to the original (put page 12 before page 3) or duplicate a page by listing its number more than once. Everything happens locally with pdf-lib — the file is copied and reassembled in your browser, never uploaded to a server.

Drop a PDF here, or

One PDF, up to 64 MB. Stays in your browser — nothing is uploaded.

How to use

  1. Drop your PDF

    Drag a PDF (up to 64 MB) onto the dropzone or click to choose one. The tool reads its page tree locally and shows the file name, page count, and size.

  2. Type your page list

    Enter a comma-separated list of page numbers and ranges, 1-based — for example 5, 1, 3-4. Pages do not need to be contiguous or in ascending order.

  3. Order it your way

    The output follows the exact sequence you typed. List page 9 before page 2 to swap their order, or type 4, 4 to include page 4 twice in the result.

  4. Extract and download

    Click “Extract pages.” The tool copies each listed page into one new PDF, in your typed order, and offers it as a single download named after the original file.

Frequently asked questions

How is this different from splitting a page range?
Splitting a range (the /pdf/split/ tool) pulls one contiguous block of pages — From 3 To 7 — and keeps the original order. This tool accepts any combination of individual pages and ranges in one go, such as 1, 3, 5-8, 12, and does not require them to be contiguous or sorted.
Can I put the pages in a different order than the original?
Yes. The output order matches the order you type, not the order the pages appear in the source. Typing 9, 2, 5 produces a three-page PDF with original page 9 first, page 2 second, and page 5 third.
Can I include the same page twice?
Yes. Listing a page number more than once — for example 3, 1, 3 — duplicates that page in the output. Each occurrence is copied independently, so both copies are full, complete pages, not references to the same one.
Is there a limit on how many pages I can pull out?
The combined output is capped at 200 pages per file, counting duplicates. This is a safeguard against accidental runaway input (like an unbounded range) rather than a technical limit — most real selections are far smaller.
Does my PDF get uploaded anywhere?
No. The file is opened and reassembled entirely with JavaScript (the pdf-lib library) in your browser. There is no server request carrying your document, so Convertitive never sees its contents.
Do the extracted pages lose any quality?
No. The tool copies the original page objects — fonts, images, and vector drawing instructions — directly into the new file rather than rendering them to pixels. The result is exactly as sharp as the source, regardless of how many times a page is duplicated or reordered.

About

How arbitrary page selection works

pdf-lib's copyPages(sourceDoc, indices) takes a source document and an array of 0-based page indices, and returns copied page objects in exactly the order the array specifies — the indices do not need to be sorted, contiguous, or unique. This tool parses the typed spec (page numbers and ranges, 1-based as PDF readers show them) into that index array, preserving order and duplicates precisely as entered, then hands it straight to copyPages. Each returned page is added to a fresh document in sequence. Because the mapping from typed spec to index array is direct and order-preserving, “5, 1, 3-4” becomes indices [4, 0, 2, 3] and produces a four-page PDF in exactly that sequence — no separate reorder or dedupe step is needed.

A real scenario: assembling a packet from a reference document

Say you have a 300-page product reference manual and need to send a client a focused packet: the cover page, the two sections relevant to their contract, and the pricing appendix at the end — in presentation order rather than the manual's internal order. With this tool you type something like 1, 42-49, 118-123, 287-290 once, and get a single clean PDF with exactly those pages, in that sequence, ready to send. Doing the same thing with a general-purpose PDF editor usually means several manual extract-and-merge steps; here it is one pass, and because it runs client-side, an internal or confidential manual never has to leave your machine to be repackaged.

Sources & references

Authoritative references behind the math, constants, and tables on this page. Verified by Buğra Sözeri on the dates shown and re-checked at every deploy.