Private PDF workflow

How to extract pages from a PDF without uploading it

To save selected pages as one new PDF, choose the page numbers, copy those pages into a new document, and verify the downloaded page count and order. A browser-based extractor can do that in local memory, but “PDF extractor” can also mean text, image, or attachment extraction—three different jobs.

Published July 11, 2026Based on a public benchmarkMaintained by Michael Joiner
Fastest route

Open the free PDF page extractor, choose the PDF, enter a page selection such as 1,3-5,9, and download the new PDF. The document bytes are processed in your browser and are not sent to GoPDFConverter for page extraction.

First decide what “extract from PDF” means

You wantCorrect operationOutputUse this
One page or selected pagesCopy PDF pagesOne smaller PDFExtract PDF Pages
Every page as a separate fileSplit the PDFMultiple PDFs in a ZIPSplit PDF
Selectable wordsExtract the text layerTXTPDF to Text
Words from scanned pagesRun OCRSearchable PDF or TXTOCR PDF
Whole pages as imagesRender the pagesPNG or JPG filesPDF to PNG or PDF to JPG
Files attached inside the PDFExtract embedded attachmentsOriginal attached filesUse an audited desktop PDF tool; GoPDFConverter currently inspects but does not export attachments

This distinction matters because rendering a page to PNG is not the same as recovering an embedded image, and extracting visible text is not the same as saving an attached spreadsheet. A useful PDF extractor should name the output rather than hide several unrelated operations behind one button.

How to save one page or a range as a new PDF

  1. Open the page extractor.

    Use Extract PDF Pages. No account or email address is required.

  2. Choose the source PDF.

    The browser reads the file into memory and shows page previews when the preview renderer supports the document.

  3. Enter one-based page numbers.

    The first visible page is page 1. Use commas for separate pages and a hyphen for an inclusive range.

  4. Create the output.

    The selected pages are copied into one new PDF in ascending page order. The source PDF remains unchanged.

  5. Open and verify the download.

    Check the page count and order. If forms, bookmarks, attachments, metadata, encryption, or digital signatures matter, inspect those separately because they are document-level features rather than page artwork.

Page-range examples

4

Creates a new PDF containing only page 4.

3-7

Creates a five-page PDF containing pages 3, 4, 5, 6, and 7.

1,3-5,9

Creates a five-page PDF containing pages 1, 3, 4, 5, and 9.

Repeated entries are deduplicated, invalid page numbers are ignored, and the current parser returns valid pages in ascending order. That means 5,2 produces pages 2 then 5, not the typed order. Use the Organize PDF tool when you need a custom output order.

What survives page extraction?

The current browser tool uses PDF-LIB to copy selected PDF page objects into a newly created PDF. It does not rasterize the page, so text and vector artwork stay as PDF content and images are not deliberately recompressed. That is more precise than saying the output is “identical.”

FeatureWhat to expectWhat to verify
Page artworkCopied as PDF page content rather than rendered to an imageVisual appearance, clipping, rotation, and page size
Text and vectorsNormally remain text/vector contentSelection, search, fonts, and accessibility reading order
Raster imagesNot intentionally downsampled or JPEG-recompressedEffective image PPI and transparency
Bookmarks and outlinesDocument-level structures may not be rebuilt for the new page setBookmark destinations and table-of-contents links
Forms and annotationsPage annotations may copy, while form relationships can be document-wideField appearance, editability, calculations, and submit actions
Attachments and metadataNot promised as part of page copyingEmbedded files, title, author, XMP, and custom properties
Digital signaturesA newly created PDF is not the originally signed byte sequenceAssume prior certificate signatures need separate validation
Password protectionThe new PDF is not automatically re-encryptedReapply protection if authorized and required

Evidence for the tested page-extraction case

The public browser PDF tools accuracy benchmark asked the production extractor to pull pages 2, 5, and 9 from a deterministic stress PDF. Independent output checks confirmed three pages in the requested ascending order. The downloadable CSV records the test case, expected result, and SHA-256 output checksum.

That evidence supports the named fixture and production release. It does not prove that every unusual PDF, encrypted file, form, portfolio, or signed document will preserve every document-level feature. The site’s testing methodology explains why opening a download is not enough: page count, order, geometry, text, images, and security behavior need independent checks.

Extraction vs splitting, deleting, and cropping

OperationWhat changesChoose it when
Extract pagesCopies chosen pages into one new PDFYou need one smaller document made from selected pages
Split PDFCreates several output PDFsYou need separate files per page or range
Delete pagesCopies everything except the pages you removeMost of the document should stay
Crop PDFChanges the visible page boundaryYou need to hide margins or part of a page, not remove whole pages

Frequently asked questions

How do I extract one page from a PDF?
Choose the PDF, enter the one-based page number such as 4, and download the new one-page PDF. The original file is not modified.
What does PDF extractor mean?
It can mean saving selected pages as another PDF, extracting selectable text, running OCR on a scan, rendering pages as images, or saving embedded attachments. Those jobs produce different outputs and need different tools.
Does extracting pages reduce PDF quality?
The current page extractor copies PDF page content instead of rendering it into pixels, so it does not deliberately downsample pages or recompress their images. Document-level features may still change or be omitted in the new PDF.
Can I preserve a custom page order?
The current range parser returns valid pages in ascending order. Extract first, then use Organize PDF if the final pages need a different order.
Can I extract pages from a password-protected PDF?
The current extractor does not ask for a PDF password. If it cannot open the file, unlock it first only when you know the password and have permission to modify the document.

Choose the exact extractor you need

Extract selected PDF pages

Create one new PDF from a single page, an inclusive range, or a combination of page numbers.

Extract selectable text

Save the existing PDF text layer as a TXT file without converting page artwork.

Extract text from scans

Use OCR when the PDF pages contain photographed or scanned words rather than a text layer.

Extract pages as PNG images

Render complete pages at a chosen DPI when an image—not another PDF—is the required output.