Skip to content
BabaPDF
Optimize

Compress PDF

Shrink file size while keeping it readable.

loading tool…

processed in your browser · 0 KB uploaded
Compress PDF · overview

What Compress PDF does

Compress PDF shrinks file size by re-encoding the photographs embedded in a document at a lower quality and resolution, using your browser's own image encoder.

PDF file size is almost always photographs. A hundred pages of text might be under a megabyte; a single high-resolution scan can be ten. So compressing a PDF really means compressing the images inside it, and how much you gain depends entirely on how many there are.

This tool finds the JPEG images embedded in your document, redraws each one smaller and at reduced quality through a canvas, and swaps it back in — but only if the new version is genuinely smaller. Afterwards it tells you exactly how many images it recompressed, so you can tell the difference between "this worked" and "there was nothing here to compress".

Step by step

How to use Compress PDF

  1. Upload the PDF

    Drop in your file. Its current size is shown so you have a baseline.

  2. Pick a compression level

    Low keeps images close to original quality; High shrinks hardest with visible quality loss. Medium is a reasonable default.

  3. Compress

    Click Compress. Each embedded photo is re-encoded on your device.

  4. Check the result

    The new size is shown next to the original, along with how many images were actually recompressed.

Details & limits

Compress PDF: details and limits

  • Effectiveness depends on how image-heavy the PDF is. A photo-heavy scan can shrink dramatically; a text-and-vector document like a contract has almost nothing to compress and may barely change. The tool says which happened rather than pretending.

  • Text and vector graphics are never touched, so they stay perfectly sharp at any compression level.

  • The three levels are specific rather than vague. Low re-encodes each image at JPEG quality 0.82 and caps its longest side at 2200 pixels; Medium uses 0.65 and 1600; High uses 0.42 and 1100. An image already smaller than the cap is never enlarged to meet it.

  • Only images stored as plain JPEG — /DCTDecode, in PDF terms — are re-encoded. Other encodings are left alone rather than risking a corrupted or worse-looking result, so a PDF full of PNG-style images can report zero recompressed and be working correctly.

  • A JPEG with a transparency mask, or in CMYK or another print colour space, is also left untouched — and counted, so the tool tells you. Re-encoding goes through your browser's canvas, which would flatten the transparency and mangle CMYK colours; we tested a magenta CMYK scan and it came back green, which is why those images are skipped instead.

  • An image is only replaced if the recompressed version is actually smaller — compression never makes part of your file bigger.

Compress PDF · help

Compress PDF: frequently asked questions

BabaPDF finds photos embedded in your PDF and re-encodes them at a lower quality using your browser's built-in image tools, then saves the document back with only those image streams changed — form fields, bookmarks and metadata stay put. Everything happens on your device.

Go deeper

Guides

Next steps