How to compress a PDF
- Drop one or more PDFs onto the drop zone above.
- Choose a compression level:
- Lossless – optimises the file structure and keeps text, links and vector graphics exactly as they are.
- Recommended – re-renders every page as a JPEG image at 150 DPI.
- Strong – re-renders every page at 100 DPI for the smallest file.
- Click Compress PDF and wait while each file is processed. The result shows the new size and how much smaller it is than the original.
- Download the compressed file. With several PDFs, Download all saves them together as a ZIP archive.
Compression is performed by pdf-lib and pdf.js inside your browser, so confidential documents are never uploaded anywhere.
What makes a PDF large?
Most of the weight in a typical PDF comes from a few sources:
- Embedded images – scans and photos stored at high resolution are usually the biggest contributor. A page scanned at 600 DPI holds far more pixels than any screen needs.
- Embedded fonts – full font files are sometimes included even when only a few characters are used.
- Structural overhead – unused objects, old revisions left behind by incremental saves, metadata and uncompressed streams.
The two families of modes attack these in different ways.
| Mode | What it does | Text selectable? | Best for |
|---|---|---|---|
| Lossless | Re-saves the document structure, packing objects and streams more efficiently | Yes | Digital documents, forms, files with links |
| Recommended | Draws each page as a 150 DPI JPEG image and builds a new PDF from those images | No | Scans and image-heavy files for email |
| Strong | Same as Recommended at 100 DPI | No | Getting under a strict upload limit |
Lossless savings vary a lot. A file that was already well optimised may shrink only slightly, while one saved many times by an editor can lose a noticeable share of its size. The image-based modes usually produce much bigger reductions on scanned documents, but they may not help a short, text-only PDF, because a page of text is very compact while a picture of that page is not. If the compressed version would not be smaller, the tool keeps your original file and marks it as already optimised.
Things to know before you compress
- Recommended and Strong remove the text layer. The result looks the same but you can no longer search, select or copy text, and links stop working. Keep the original if you may need those.
- At 100 DPI small print can look soft when zoomed in or printed. Use Recommended if the document will be printed.
- Password-protected or encrypted PDFs are not supported. Unlock it first with Unlock PDF (you need its password), then come back.
- If a PDF is large because it contains photos you added yourself, shrinking the photos with Compress Image before creating the PDF often gives better quality for the same size.
Need only part of the document? Removing pages you don't need with Split PDF is another simple way to cut the size.
Frequently asked questions
Why can't I select text after compressing?
Recommended and Strong modes turn every page into an image. That gives large savings on scans but removes the text layer. Use Lossless to keep selectable text.
Why did my PDF barely shrink?
In Lossless mode the content is left untouched, so a file that is already efficient has little to remove. Try Recommended if the document is mostly scans or photos.
Why did I get my original file back?
When compression would not reduce the size, the original is kept so you never end up with a bigger file. This is common for text-only PDFs in image-based modes and for files that are already well optimised.
Can I turn the compressed pages back into editable text?
Not from the image-based output, since it no longer contains text. Convert the original instead, for example with PDF to Word.