Compress PDF

Reduce PDF file size by re-encoding the images inside the document. Three compression levels, page count and per-image statistics, and the text layer is never rasterised.

Upload PDF

📄

Drag and drop your PDF here, or click to select (max 25 MB)

Result

🗜️

Your compressed PDF will appear here

Upload a PDF and pick a compression level

About PDF Compression

  • • Unlike our other PDF tools, this one processes your file on our server — it is discarded straight after
  • • Only embedded images are re-encoded; text and vector content are never rasterised
  • • Text stays selectable and searchable, and stays sharp at any zoom level
  • • PDFs with no raster images will barely shrink — that is expected, not a failure
  • • CMYK images are skipped so print colours are not shifted to sRGB
  • • Password-protected PDFs must be decrypted before they can be compressed

About Compress PDF

This PDF compressor reduces file size by re-encoding the raster images embedded in a document. Each JPEG image inside the PDF is extracted, optionally downsampled, and re-encoded at a lower quality, then written back into the same document structure. Text, fonts, and vector artwork are left completely untouched, so the document remains searchable and stays sharp at any zoom.

Why use a Compress PDF?

Scanned documents are frequently far larger than they need to be, because scanners save pages as high-resolution photographs. That routinely pushes a handful of pages past an email attachment limit or a government upload form's cap. Compressing the embedded images brings the file back under those limits without turning the document into a picture.

Who is it for?

Useful to anyone blocked by a file size limit: applicants uploading scanned certificates to a portal that caps at 5 MB, accountants emailing receipt bundles, students submitting scanned coursework, and administrators archiving paperwork where storage volume matters.

How to use the tool

1

Drag your PDF onto the drop zone, or click to browse for a file

2

Pick a compression level — Light preserves the most detail, Strong shrinks the most

3

Click Compress PDF and wait for processing to finish

4

Check the reported ratio and how many images were re-encoded

5

Open the preview to confirm the quality is still acceptable for your purpose

6

Download the compressed file, or try a different level if the result is too soft

Frequently Asked Questions

How do I compress a PDF?

Upload your .pdf file, choose one of the three levels, and click Compress PDF. Light re-encodes embedded images at quality 85 and caps them at 2400 pixels wide; Balanced uses quality 70 at 1600 pixels; Strong uses quality 50 at 1000 pixels. The tool reports the original and final size, the reduction percentage, and how many images it re-encoded, so you can judge whether to try a different level before downloading.

Is my PDF uploaded to a server?

Yes — unlike our other PDF tools, this one cannot run in your browser. Compression requires re-encoding the images inside the document with libraries that need a server runtime. Your file is sent over TLS to our processing pipeline and returned to your browser, and it is discarded immediately after processing — we don't store, log, or share it. For confidential documents, compress locally instead with Ghostscript: `gs -sDEVICE=pdfwrite -dPDFSETTINGS=/ebook -o out.pdf in.pdf`.

Why did my PDF barely get smaller?

Because it probably contains little or no raster imagery. A PDF's size is dominated by what it embeds, and text, fonts, and vector graphics are already stored efficiently — there is very little left to reclaim. This tool reports how many images it found; if that number is zero, the file was already about as small as it can be without destroying content. Text-heavy PDFs exported from Word or LaTeX typically fall into this category.

Does compression make the text blurry or unsearchable?

No. Only embedded raster images are re-encoded — text is left exactly as it was, so it stays selectable, searchable, and perfectly sharp at any zoom. This is the key difference from compressors that rasterise every page into a picture: those achieve bigger reductions but destroy the text layer, break screen-reader accessibility, and make the document unsearchable. If your PDF is a scan, the 'text' is already an image and will soften as you compress harder.

Which compression level should I choose?

Start with Balanced, which is a good default for scanned paperwork intended for screen reading. Choose Light when the document will be printed or contains detail that matters, such as photographs or fine diagrams. Choose Strong only when you must hit a hard size limit and legibility on screen is all that counts — at quality 50 and a 1000 pixel cap, scanned text remains readable but visibly soft. You can re-run a different level without re-uploading a fresh copy.

Can it compress a password-protected PDF?

No. A PDF protected with a user password has encrypted content streams, so its images cannot be read or re-encoded, and the tool returns a clear error rather than a broken file. Remove the protection in a PDF reader you have the password for, save an unprotected copy, and compress that. Note that compression does not preserve any encryption or permission flags the original carried, so re-apply them afterwards if you need them.

Why were some images skipped?

Two cases are skipped deliberately. Images stored in CMYK are left alone, because re-encoding them converts to sRGB and would shift every colour on the page — unacceptable for print-intended documents. Images that come out no smaller after re-encoding are also left alone, which happens when they were already compressed aggressively. Images stored with non-JPEG filters are not touched either, since rebuilding their raw sample data risks corrupting the page silently.

Is there a file size limit?

Uploads are capped at 25 MB, which covers the great majority of scanned documents. If your file exceeds it, split it into pieces with [Split PDF](/tools/split-pdf/) — which runs entirely in your browser — compress each piece, then recombine them with [Merge PDF](/tools/merge-pdf/). Very large documents are usually scans at unnecessarily high resolution, so compressing in parts often brings the recombined total comfortably under the original.

Share This Tool

Found this tool helpful? Share it with others who might benefit from it!

💡 Help others discover useful tools! Sharing helps us keep these tools free and accessible to everyone.

Support This Project

Buy Me a Coffee