How can I send only some pages of a PDF without sending the rest?
Extract just those pages into a new PDF and send that file instead of the original. In our test, extracting pages 2 and 5 of a six-page sample gave a two-page file with none of the other pages' text and none of the original's Author or Subject. Check your own file before you rely on it.
How do I extract pages?
- Open the Split PDF page and choose your file.
- Pick "Keep only these pages" and type the pages, for example
2, 5or1-3, 8-. A range ending in a dash means "to the last page". - Press Extract pages and download the new PDF.
Pages are counted from the first page of the file, not from the numbers printed on them. Pages come out in the order you type them. The original file is not changed: the tool reads it in your browser and writes a separate new file.
What did we test?
We made a six-page sample PDF with invented text. Each page carries a unique reference word, and the file has document properties filled in: Title "Service agreement (sample)", Author "Jane Example", Subject "Confidential sample". We extracted pages 2 and 5 with the Split PDF tool on osmanthdocs.com, downloaded the result, and read it with a PDF library.
| Original | Extracted file | |
|---|---|---|
| Pages | 6 | 2 |
| Text on page 1 | "Section 1 ... ALPHA-01" | "Section 2 ... BRAVO-02" |
| Text on page 2 | "Section 2 ... BRAVO-02" | "Section 5 ... ECHO-05" |
| Title | Service agreement (sample) | empty |
| Author | Jane Example | empty |
| Subject | Confidential sample | empty |
| Creator and Producer | Example Word Processor, ReportLab | pdf-lib |
| File size | 4,342 bytes | 1,699 bytes |
We searched the extracted file's text for words from the other four pages (ALPHA-01, CHARLIE-03, DELTA-04, FOXTROT-06) and for the original author's name. None was there. We also searched the raw file bytes for the Author and Subject strings, and they were absent.
In a second extraction, with the browser's network log recording, we saw only GET requests for the site's own files and no request to an API, so the file was not uploaded.
What did our test not check?
Our sample is a simple generated file. Real documents can carry more, and we did not test these:
- XMP metadata. Besides the document information fields, a PDF can hold metadata in a separate XMP stream, which is XML. The PDF Association's notes on ISO 32000-2 show it as its own object, separate from the information dictionary, and Adobe describes embedding XMP in PDF and reconciling it with other metadata. Our sample had no XMP stream, so we cannot say what happens to one.
- Attachments, form data and comments. We did not test files that contain them.
- What is visible on the page. An extracted page still shows everything that was on it, including text hidden under a black box that was only drawn over it.
For anything sensitive, open the new file, look at its properties, select all text on every page, and try a search for words that should not be there.
Can I check that the page did not upload my file?
Yes. Open the browser's Network panel before you add the file, clear it, extract the pages, and look for requests that go out when you press the button. Chrome's documentation describes the Network panel, its Clear button and the Payload tab that shows the data sent with a request. Our guide to merging PDFs safely has the full steps.
What happens to your file?
Split PDF runs in your browser and does not upload your file. The conversion tools on this site are different: your PDF is uploaded to our servers, where open-source software processes it. It is not sent to any third-party AI or document API. The uploaded PDF is deleted as soon as processing finishes, the results stay for 60 minutes, and files can be up to 50 MB and 100 pages. See the privacy page.