Extracting PDF pages for free means building a brand-new PDF that contains only the pages you select from a source document, and the Extract PDF Pages tool at Lizely does this in your browser with a hard limit of 50 MiB and 500 pages per source file, accepting entries like 5 or 1-3 in the exact output order you specify. The browser reads the original bytes locally, the pdf-lib library copies each requested page object into a fresh document, and you download the result without ever uploading the source. What the parser is strict about is the syntax: one-based page numbers, comma or whitespace separators, inclusive ascending ranges, and a deterministic first-occurrence rule for duplicates. Documents protected by a password cannot be opened by the tool, and features that depend on objects outside the selected pages — digital signatures, AcroForm fields, bookmarks, and dynamic XFA forms — will not survive the extraction in their original state. The rest of this guide covers the exact steps, the syntax for picking pages, the limits you need to respect, and what to verify in the new PDF once it is ready.
What Extracting Pages From a PDF Actually Does
Extracting pages produces a separate, downloadable PDF that contains only the pages you choose. The original file stays on your computer and is not modified. Inside the tool, the browser reads your source bytes through the File API, the pdf-lib PDFDocument class loads the document in memory without bypassing encryption, and a fresh PDFDocument is created. Each requested page object is copied across, carrying its media box, crop box, page graphics, ordinary page resources, and any 90-degree rotation. The new file is then saved to a temporary local Object URL that the download button exposes.
This is the right operation when you want to pull a single chapter out of an eBook, send only the signed appendix of a contract, collect scattered reference pages into a short handout, or assemble a custom packet from a long report. It is not the same as splitting a PDF into many single-page files; if you need one PDF per page, use a dedicated splitter. It also differs from deleting pages, because the original stays untouched and a separate file is produced rather than a trimmed version of the original.
How to Extract PDF Pages for Free in Your Browser
- Open the Extract PDF Pages tool on Lizely in a current desktop browser such as Chrome, Firefox, Safari, or Edge.
- Click the file picker and choose an unencrypted PDF stored on your computer. The file must be no larger than 50 MiB and must contain between 1 and 500 pages.
- Wait for the browser to finish reading the source.
- Type the page numbers or ranges you want into the selection field, in the output order you want. Use commas or whitespace to separate tokens and a hyphen for inclusive ascending ranges, for example 5, 1-3 or 2, 4, 6-8.
- Click the extract button. The tool copies the matching page objects into a new PDF document and exposes a temporary download link.
- Read any duplicate-page notice if it appears. The first occurrence of every page is kept and later occurrences are ignored so the result still matches your typed order.
- Click the download link to save the new PDF to your device. Use the local source link if you want to compare the original and result side by side.
- Close the page when you are done to release the temporary source and result URLs held by the browser.
Page Selection Syntax at a Glance
The parser is strict on purpose so a stray character cannot quietly change your output. The table below shows how common inputs map to pages in the output for a sample 10-page document. Pages are numbered the way a PDF viewer shows them, starting at 1.
| Input you type | Pages in the output | Why |
|---|---|---|
| 5 | 5 | A single page token. |
| 1, 3, 5 | 1, 3, 5 | Comma-separated single pages. |
| 1-3 | 1, 2, 3 | Inclusive ascending range. |
| 5, 2-3 | 5, 2, 3 | Token order controls output order. |
| 3, 1-3 | 3, 1, 2 | Second occurrence of page 3 ignored. |
| 5-3 | Rejected with an error | Descending range is not silently reversed. |
| 2.5 or -1 | Rejected with an error | Decimals and negatives are not allowed. |
| 1, 12 | Rejected with an error | 12 is outside the 10-page source. |
To see the parser at work on one concrete expression, take the input 1-3, 5, 7-9 against a 10-page source. Reading left to right, the tokens expand to the sequence [1, 2, 3, 5, 7, 8, 9], which is also the output page count: 7 pages in the new PDF. Because none of those pages repeat, the duplicate-occurrence rule does not change anything in this case.
File Limits and Features That Will Not Survive
Knowing the boundaries up front saves a failed run. The limits below are enforced as hard cutoffs — a file at the boundary is accepted, and a larger file or longer document is rejected rather than trimmed.
| Constraint | Value or behavior |
|---|---|
| Maximum file size | 50 MiB (52,428,800 bytes); larger files are rejected. |
| Maximum page count | 500 pages in the source; a longer document is rejected. |
| Encryption | Password-protected PDFs are not opened; encryption is never bypassed. |
| Selection field length | Up to 4,000 characters; over-limit input is rejected rather than truncated. |
| AcroForm fields | May lose interactivity because fields depend on objects outside selected pages. |
| Digital signatures | Do not remain valid after extraction; the output is a different file. |
| Bookmarks, outlines, links | Cross-page destinations and document-level structures may break. |
| Dynamic XFA forms | Not a supported preservation target in pdf-lib. |
| Damaged or unsupported files | Can fail to load. |
One related point worth flagging is output file size. Selected page resources such as fonts and images are copied into the new document, so the result can still be large even when you extract a single page from a resource-heavy source. File size does not scale in direct proportion to page count, and browser memory availability will vary by device and tab workload.
Why a Browser-Based Extractor Protects Your File
Because every step runs locally, your source PDF never leaves the device. The browser validates the PDF MIME type or extension and the 50 MiB byte cap, then pdf-lib loads the source without ignoreEncryption so encrypted files surface as an error instead of silently producing a wrong output. The parser builds a zero-based index list that the production function passes straight to pdf-lib, and the resulting bytes are exposed only through a temporary Object URL held by the current tab.
Two temporary URLs do exist while the page is open: one for the source preview and one for the result. They are released when you change files, run a new extraction, or close the page, and job-level guards prevent an older asynchronous load or save from overwriting newer state. A real integration test for the production function builds a source with mixed page sizes, runs it through the extractor, reopens the saved bytes, and verifies the output count, order, dimensions, and page identity. That is why a correct result has the pages you typed, in the order you typed them, with the page sizes preserved.
After You Download the New PDF
Before you treat the extracted file as final, open it in the viewer your recipient will use and check three things: that every listed page is present, that they appear in the order you specified, and that any field, link, or annotation you depend on still behaves correctly. Keep the original PDF alongside the new one until you have confirmed the output, and switch to a specialist PDF editor when the document depends on digital signatures, XFA, accessibility tags, portfolios, attachments, or regulated archival conformance.
If you also need a related operation, splitting the source into many single-page files is a job for the Split PDF tool, while combining several PDFs into one longer document is handled by Merge PDF. Both follow the same browser-only approach, so neither requires uploading your files.