Sycago
All toolsReport an issue ↗

PDF tools / FREE ONLINE TOOL

Extract PDF Pages

Copy selected pages from an unencrypted PDF into a new document while keeping the source intact.

No signupNo API keysLocal processing

YOUR WORKSPACE

Extract PDF Pages

How to use ↓

Add your files

Drop files here, or choose from your device.

Unencrypted PDF · 25 MB per file · 100 MB total
One input file · Device memory may impose lower limits.

Ready. Your inputs stay in this tool's browser session.

Sycago Extract PDF Pages copies chosen pages from an unencrypted PDF into one new document. Select a single page, several ranges, or a deliberate sequence, and download a focused PDF while keeping the source intact. The requested order is preserved, which makes extraction useful for both short excerpts and carefully arranged reading packets. A selection such as 8,2-4 produces source page 8 followed by pages 2, 3, and 4. The important decisions are which physical pages belong in the excerpt, how they should be ordered, and what context readers need to interpret them correctly.

Choose the smallest complete excerpt

A useful excerpt contains enough information for its purpose without forcing the reader through the whole source. A project reviewer may need the summary and findings, a technician may need the relevant procedure, or a colleague may need a particular appendix. Start by naming the reading task. Then identify the pages that answer it and any nearby pages that explain important definitions, units, assumptions, or qualifications.

The smallest possible selection is not always the best selection. A chart page without its footnote can be misleading. A procedure without the preceding safety conditions can be incomplete. A table may begin on one page and continue on the next. Open the source around every candidate page and check whether the material depends on content just outside your initial choice. A deliberate extra page can save the recipient from misunderstanding the excerpt.

Extraction creates one new PDF. If you need several separate section files, use a split workflow instead. If you want to retain almost everything except a few unwanted pages, a deletion workflow may be easier to describe and review. If every page must appear once in a new sequence, a full reordering workflow provides a clearer completeness check. Choose the operation that matches the structure you actually need.

Use physical page positions rather than printed labels

The first physical page in the PDF is position 1. This remains true if the cover has no printed number or the opening pages use Roman numerals. A body page labelled 12 may be physical page 16 after four preliminary pages. Use a PDF reader to identify the actual positions of the pages you want, then enter those positions in the extraction field.

Make a short source-to-output map for anything more complicated than a single continuous range. Record the source position, its visible heading, and where it should appear in the excerpt. For example, source page 8 might be the summary, source pages 2-4 the method, and source page 10 a reference chart. This map catches mistakes that a bare list of numbers cannot, because you can compare the downloaded pages with meaningful headings.

Check the physical count before entering a range endpoint. A request for page 31 cannot work in a 30 page source even if the document's printed footer happens to show 31. If a PDF contains blank scanner pages, those are still physical pages. Count them when identifying positions, or prepare a cleaned copy first and then remap your selection against that new source.

Express ranges and requested order clearly

SelectionOutput sequenceCommon use
4Source page 4A single reference page
2-5Source pages 2, 3, 4, 5A continuous section
1,6-8Source page 1 followed by 6, 7, 8A cover plus a section
8,2-4,10Source pages 8, 2, 3, 4, 10A summary-first reading packet

Ranges include both endpoints. The selection 2-5 contains four pages. Commas separate the requested pieces, and extraction preserves their requested order. This differs from a selection process that silently sorts everything back into source order. If you deliberately put page 8 first, verify that it is the first output page after download.

Use a clear, nonoverlapping selection unless there is a specific reason to do otherwise. Avoid ambiguity from repeated positions, stray punctuation, or pasted headings. Every page must be within the source's physical count. If the interface rejects your entry, simplify it to one valid page and build up the selection carefully. Keep the intended output sequence written down so a syntax correction does not accidentally change the reading order.

Create and review the extracted PDF

  1. Add the unencrypted source PDF.
  2. Identify the required physical page positions.
  3. Enter single pages and ranges in the desired output order.
  4. Check the expected number of selected pages.
  5. Run extraction and download the new PDF.
  6. Open the downloaded file and compare every output page with your plan.

For a multi-part selection, review both membership and order. Membership asks whether all the needed source pages are present. Order asks whether those pages appear in the intended sequence. A four page result can have the right length while containing the wrong four pages, or the right pages in a confusing order. Use headings and content landmarks alongside the page count.

Save the excerpt with a name that identifies its purpose, such as project-findings-excerpt.pdf. Keep the full source separately. If the source changes, create a new extraction from the updated file rather than assuming that the same page positions still refer to the same content. An inserted cover or new appendix can shift positions throughout a document. Recheck the map whenever you change the source version.

Worked example: a summary-first review document

Suppose a ten page report places its detailed method on physical pages 2-4, a concise summary on page 8, and a reference table on page 10. A reviewer wants the summary first, followed by the method and table. Enter 8,2-4,10. The expected result has five pages: summary, three method pages, and the reference table. The original report remains in its original order.

Before extraction, read the summary's references. It may say that definitions appear on page 5 or that the method depends on an assumption explained on page 1. Decide whether those references matter to the review. If they do, include the relevant context or prepare an explanatory note outside the extraction workflow. Page selection cannot rewrite references embedded in the source text, so the excerpt should be understandable despite them.

After downloading, inspect the first page to confirm that the summary really is first. Check that the three method pages remain consecutive and complete, and that the final table includes its legend and units. If the table continues onto another page, revise the selection. The goal is a useful review document, not merely a technically correct five page output with incomplete information.

Worked example: share one procedure from a manual

A 75 page manual contains the required procedure on physical pages 31-35. At first it looks as though 31-35 is all you need. Reading the surrounding pages reveals that page 30 defines the equipment configuration and page 36 contains the final verification checklist. A complete excerpt should therefore include 30-36, producing seven pages. That extra review prevents the recipient from using instructions without their setup conditions or final checks.

Enter the complete range and download the new PDF. Check the opening configuration page, the transition into the instructions, and the final checklist. If the source uses printed page labels that differ from physical positions, the visible numbers in the excerpt will still show the source numbering. Extraction does not automatically replace them with 1 through 7. This can be useful for traceability, but tell readers which manual version the excerpt came from when that context matters.

Keep the original manual available. An excerpt can become outdated independently of its source if people continue forwarding it after a revision. Use a descriptive filename and an appropriate version reference in your distribution process. This tool copies the selected pages; it does not monitor later source updates or guarantee that a separated procedure remains current.

Preserve meaning when pages leave their source

Documents rely on context in subtle ways. A figure caption may define a scale, a footnote may qualify a result, and a heading on the previous page may identify the organization or reporting period. Before extracting, read the boundaries as though you had never seen the complete source. Ask whether the first page explains what the excerpt is, whether abbreviations are defined, and whether the last page ends a complete thought.

Visible cross-references do not update themselves. A sentence saying see page 14 still refers to the source's numbering even if the extracted PDF has only three physical pages. Internal links may also depend on destinations outside the excerpt. If those references are necessary, include the referenced pages or provide a clear accompanying explanation through your normal communication process. The extraction operation does not edit source prose.

Be careful when rearranging pages. Putting a conclusion before its supporting evidence can make a packet easier to scan, but it can also change how a reader interprets it. Preserve the original source and avoid presenting a rearranged excerpt as though it were the complete original document. For records where sequence is material, keep source order unless the recipient explicitly needs a different arrangement.

Selecting pages is not the same as redacting content

Extraction can omit whole pages from a new copy. It does not remove individual names, sentences, photographs, or objects from a retained page. If a selected page contains confidential information beside the content you want to share, that information remains part of the selected page. Use a proper content removal workflow when information must be removed, then verify the prepared source before extracting.

Do not assume that a small page view or a hidden-looking area is safe. PDF structure can include annotations, forms, metadata, and other content beyond a simple screenshot of the visible page. This tool is not a certified sanitization process. When your task has specific disclosure requirements, follow the appropriate review procedure and examine the resulting PDF with tools suited to that requirement.

Local browser processing means the source is not sent to an external conversion API, but human selection still matters. A page selection error can disclose an unwanted record even when processing stays local. Review the exact downloaded excerpt, including headers, footers, and continuation pages, before sharing. Keep separate filenames for trial outputs and the final reviewed copy to reduce the chance of sending the wrong version.

Check specialized PDF features after extraction

Page extraction can preserve ordinary PDF page content without turning every page into a bitmap. It should not be treated as a guarantee that bookmarks, interactive forms, document attachments, navigation, and other document-level structures are reproduced exactly. Inspect the features that matter in the new PDF. A page that looks right may still have a link or field behavior that differs from the source.

Rewriting a PDF generally invalidates existing cryptographic signatures. A visible signature appearance may remain without providing valid verification for the extracted document. Preserve the original signed source when signature checking is required. If a recipient needs the signed complete record, an excerpt is not an equivalent substitute merely because the signature graphic is visible on one selected page.

Password protected PDFs are unsupported. Damaged files and specialized forms may also fail to read or copy reliably. Obtain an authorized ordinary export from the source application when necessary. Extraction does not provide OCR, PDF compression, DOCX conversion, or repair of a malformed source. If scanned text was not searchable before extraction, the operation does not create a text layer.

Respect input limits and device resources

A source PDF can be up to 25 MB. Shared PDF workflows allow up to 20 files and 100 MB combined, with a maximum of 400 total source pages. This extraction workflow uses one source, and that source must fit the applicable limits even when you select only a few output pages. Selecting one page from an oversized source does not make the source disappear from the loading step.

The browser loads and writes the document locally using bundled PDF functionality. A permitted file can still stress a device with limited available memory, especially when it contains large scans or complex page content. Close unrelated demanding tabs, keep your selection plan, and retry with a suitable prepared source if the browser cannot complete the operation. Avoid repeatedly starting overlapping jobs.

The new file is a separate download. It does not overwrite or rearrange the source on disk. Check the output size if the recipient has a limit, but do not assume that selecting fewer pages guarantees a particular byte size. Some pages contain much more data than others. The tool does not offer a separate compression operation, so source preparation and actual output inspection remain important.

Troubleshoot incorrect pages, order, or output behavior

If a selection is rejected, confirm that all numbers are physical positions inside the source and that the range syntax is clear. Test one valid page if necessary. If the downloaded pages are wrong, compare printed labels with physical positions. If the order is wrong, read the requested sequence from left to right and compare it with the source-to-output map you intended.

If the source will not load, verify that it is unencrypted and opens in another PDF reader. The tool cannot unlock a password protected file. A damaged export or specialized form document may need a fresh authorized export from its source application. Changing a filename extension does not fix structural problems inside a PDF.

If a result seems incomplete, check the selected page count, continuation pages, and the source version. A new source revision may have shifted positions. If links, fields, or signature behavior differ, compare the extracted copy with the original and decide whether the excerpt is suitable for the task. Before sharing, open the actual downloaded file and confirm that its first page, last page, and complete sequence match the plan.

Record where each excerpted page came from

A short provenance note can make a rearranged excerpt easier to review. Record the source filename and version, the requested selection, and the output position of each selected block. For 8,2-4,10, that means output page 1 comes from source page 8, output pages 2-4 come from source pages 2-4, and output page 5 comes from source page 10.

This note is useful when a reviewer asks about a statement or wants the surrounding pages. It also helps you distinguish an intentional summary-first arrangement from an accidental sequence mistake. Keep the note with your working files rather than assuming a reader can reconstruct source positions from visible page numbers alone.

When updating the excerpt, verify that the source version is still the intended one, rebuild the mapping against its actual positions, and inspect the new result. Retaining that small amount of context saves repeated investigation when documents have similar filenames or when preliminary pages change between revisions.

Frequently asked questions

What does extraction create?

It creates one new PDF containing the selected source pages. The original remains intact. Use extraction for a focused excerpt or a deliberate selected sequence, and choose splitting instead when you need several separate output PDF files.

Can I select more than one range?

Yes. Use commas to separate pages and ranges, such as 1,4-6,9. Check the physical positions in the source and calculate the expected output length. A range includes both endpoints, so 4-6 contributes three pages.

Is my requested page order preserved?

Yes. A selection such as 8,2-4 places page 8 first, followed by pages 2, 3, and 4. Review the downloaded sequence explicitly when you use this flexibility, because extraction does not simply sort every request into source order.

Do page numbers refer to printed labels?

No. They refer to physical positions in the PDF, beginning at 1. Covers, contents pages, Roman numerals, and blank scanner pages can make printed labels differ from those positions. Verify the source in a PDF reader before selecting.

Can I extract from a password protected PDF?

No. Only unencrypted PDFs are supported. The tool does not accept passwords or remove encryption. Obtain an authorized unencrypted export from the owner or source application if that is appropriate for your use.

What are the source limits?

A PDF can be up to 25 MB, with PDF operations limited to 400 total source pages. Shared workflows also cap jobs at 20 files and 100 MB combined. The source must fit applicable limits even if the output selection is small.

Will the original document be altered?

No. The operation produces a new downloaded PDF. Keep the original and use a descriptive excerpt filename. That makes it easier to correct a selection mistake, confirm source context, or rebuild the excerpt after the source changes.

Does extraction remove private text on retained pages?

No. Whole-page selection does not redact individual content. Private information on a retained page remains part of that page. Prepare a correctly redacted source through a suitable workflow when content removal is required, and inspect the result independently.

Will my pages become pictures?

The extraction copies PDF pages rather than intentionally rendering them all as images. Ordinary text and vector content can remain PDF content. Specialized document features still need checking, and a scan remains image-based unless a text layer already exists.

Does this add searchable text to scans?

No. The tool does not perform OCR. Selecting a scanned page does not recognize its text or make it editable. Keep that distinction in mind when a recipient expects to search or copy text from the excerpt.

Are existing digital signatures still valid?

Rewriting generally invalidates cryptographic signatures. A visible signature appearance is not proof that the new excerpt has valid signature verification. Keep the original signed source when verification matters and check the receiving process's requirements.

Will links and form fields survive exactly?

Do not assume every document-level feature is reproduced. Inspect relevant fields, links, and navigation in the downloaded excerpt. Links to omitted pages may no longer help the reader, and specialized forms can fail or behave differently after rewriting.

Can I compress the excerpt here?

This tool does not provide PDF compression. Fewer pages may produce a smaller document, but page content varies and no particular size reduction is guaranteed. Check the actual output against any recipient limit and prepare suitable sources separately if necessary.

Are source files sent to an external API?

No external conversion API receives the PDF for this operation. Loading and extraction happen in your browser using your device's resources. Continue to review page selection carefully and handle the downloaded excerpt according to its contents.

Why did the same range select different content later?

The source may have changed. An inserted cover, new introduction, or removed blank page shifts physical positions. Recheck your source-to-output map whenever you use a new version instead of assuming an old range still identifies the same content.

What should I verify before sharing?

Check page membership, requested order, expected count, complete tables or procedures, visible private information, and any important document features. Confirm the source version and open the exact downloaded excerpt you intend to send. Keep the full original for context.