To extract PDF pages in Adobe workflows, you can open the Organize Pages tool in Acrobat, choose Extract, and save a new file containing only the pages you select. The browser-based alternative — Extract PDF Pages — creates a new PDF but reads the original bytes, parses the requested page list, and copies the matching page objects into a fresh document inside your browser tab. The source file, selection, and output are not uploaded to Lizely, and the new PDF is offered through a temporary local download URL. The selection uses ordinary one-based page numbers, so the same "page 5" you see in Acrobat is the same page 5 you type in the tool. Because the original file and the resulting PDF never leave your device, this method also avoids the upload step that some Adobe online services require, which is useful when the source contains sensitive material or when you are working offline.
What Adobe Acrobat's Extract Pages Does
Acrobat's built-in extraction lives behind the Organize Pages tool, which you open from the right pane or the Tools center. Once the source PDF is loaded, you select a continuous or non-continuous set of thumbnails, click Extract, and Acrobat writes a new file containing only those pages. The extracted file inherits the rotation, crop, and page resources of the originals, which matches what pdf-lib-based tools do for the same selection. The Acrobat dialog also lets you choose "Delete pages after extraction," which removes the chosen pages from the source rather than copying them out. Adobe's approach runs entirely in the desktop application, so it depends on a valid Acrobat license and the original file being openable on your machine.
The online counterpart, Acrobat on the web, follows a similar flow but uploads the document to Adobe's servers for processing. That model is convenient when you only have a browser available, but it requires a sign-in and sends your source bytes over the network. When the goal is simply "produce a new PDF containing pages 5, 2, and 3 in that order," running that work in the current browser tab keeps the file local and removes the account requirement while preserving the exact order you typed.
How to Extract PDF Pages in Your Browser
- Pick a local PDF that is not password-protected, no larger than 50 MiB (52,428,800 bytes), and contains no more than 500 pages.
- Open the Extract PDF Pages tool and load the file from your device. The browser validates the PDF type and size before parsing any pages.
- Type one-based page numbers and inclusive ascending ranges separated by commas or whitespace — for example, 5, 1-3 — in the output order you want.
- Run the extraction. The browser parses the tokens, keeps only the first occurrence of any repeated page, and copies the matched page objects into a fresh PDF using pdf-lib.
- Read the duplicate-page notice if any of your tokens repeat a number, then download the new PDF from the temporary local URL. The source bytes, the selection, and the output are not uploaded to Lizely.
- Open the downloaded PDF in your usual viewer and confirm the page order, page appearance, and any links or form fields before using it.
Page Selection Syntax and What It Produces
The selection field follows a strict grammar so the same input always produces the same output. Tokens are one-based page numbers or inclusive ascending ranges; commas and whitespace separate tokens, and a hyphen joins the endpoints of a range. Token order is significant — it becomes output order — and any duplicate token after the first occurrence is ignored while being reported back to you. Descending ranges such as 5-3 are rejected, not silently reversed, and decimal values, negative values, malformed ranges, and pages outside the loaded document are rejected with an explicit error. The field accepts at most 4,000 characters, and over-limit input is not truncated or partly applied.
| Input | Pages produced (in order) | Notes |
|---|---|---|
| 5 | 5 | Single page token. |
| 1, 3, 8 | 1, 3, 8 | Comma-separated singles. |
| 3-5 | 3, 4, 5 | Inclusive ascending range. |
| 5, 2-3 | 5, 2, 3 | Token order sets output order. |
| 3, 1-3 | 3, 1, 2 | Second "3" is ignored; a duplicate notice is reported. |
| 5-3 | (rejected) | Descending range is not silently reversed. |
File Limits and What Happens at the Boundaries
The input boundary is 50 MiB, or 52,428,800 bytes, and the decoded document boundary is 500 pages. A file at either boundary is accepted; a larger file or a document that decodes to more than 500 pages is rejected, not shortened. The output can still be large because selected page resources such as fonts and embedded images may be copied into the new PDF, so file size does not necessarily scale in direct proportion to page count. Extracting one page from a resource-heavy document can retain the data that page needs to render correctly, which is why a one-page output may be larger than 1/Nth of the source. Browser memory also varies by device and tab workload, so a file that processes comfortably in one session may fail on another with less headroom.
| Boundary | Value | Behavior |
|---|---|---|
| Maximum source size | 50 MiB (52,428,800 bytes) | Accepted at the limit; larger files are rejected. |
| Maximum source page count | 500 pages | Accepted at the limit; longer documents are rejected. |
| Maximum selection length | 4,000 characters | Over-limit input is not truncated or partly applied. |
| Encryption | Not bypassed | Password-protected PDFs are rejected. |
| Output format | New PDF, vector page objects | Pages are copied, not rasterized to screenshots. |
What Extraction Preserves and What It Drops
Each copied page brings its rotation, crop box, media box, page graphics, and ordinary page resources into the new document. What does not survive cleanly is anything that depends on objects outside the selected pages: AcroForm fields, annotations, scripts, embedded files, layers, page labels, bookmarks, document outlines, cross-page destinations, and links can all break when their backing objects are left behind. Digital signatures will not remain valid after a new PDF is created, because the byte-level hash has changed. Dynamic XFA forms are not a supported preservation target in pdf-lib, so any XFA-based form should be handled in its original authoring software rather than passed through an extractor. Damaged or unsupported files can also fail to load, and the loader does not bypass encryption to peek inside a protected document.
Verifying the Output Before You Use It
Because extraction is destructive of document-wide structure by design, a quick verification pass is worth the few minutes it takes. Compare the new PDF's page count to the unique page count in your selection; they should match exactly, because the first-occurrence rule guarantees no output page appears twice. Spot-check the first and last page, then scroll through to confirm the order you asked for. Open any cross-page links and any form fields your workflow depends on. If the original was signed, the new PDF will not be signed, so re-sign it from the source if signature validity matters. If your workflow involves regulated archival conformance, accessibility structure, or XFA behavior, run the source through the appropriate specialist editor and treat this tool only as a quick extraction step.
For teams that already have an Adobe Acrobat license and need full Organize Pages fidelity, the desktop tool is still the safest path. For everyone else — anyone on a borrowed machine, behind a strict firewall, or just wanting a one-off extraction — the browser workflow described above produces a new PDF, in the order you typed, with the source bytes staying on the device the whole time.