How does PDF compression work?
Most of a PDF's size is images and embedded fonts, not text. Optimisers combine several techniques: re-encoding images at a lower resolution or quality (JPEG for photos, lossless Flate for line art), subsetting fonts to the glyphs actually used, removing unused objects, duplicate resources and metadata, and linearising the file so a browser shows the first page before the download finishes. Not every tool uses all of them. These techniques keep the text layer, links and structure; rasterising whole pages into images, the most aggressive option, turns text into pixels and loses them.
Read the guide: PDF Compression Explained
How to use Compress PDF
- Drop one or more PDFs.
- Pick a mode and, for images, the quality and maximum size.
- Click "Compress".
- Compare the sizes and download.
Compress PDF features
- Image recompression: JPEG quality, downscaling, optional grayscale
- Lossless structural rewrite: object streams, unused objects and history removed
- Page rasterisation for maximum reduction (labelled)
- Real before / after sizes and percentage
- Original kept when nothing is gained
- Batch of files with a ZIP; local processing
Compress PDF example
A scanned contract
Input:
contract.pdf — 18.4 MB, 12 pages of 300 DPI colour scans; mode: re-compress images, quality 70, max 1800 pxOutput:
contract-compressed.pdf — 2.9 MB (84% saved); notes: 12 of 12 images re-encodedFrequently asked questions about Compress PDF
Which mode should I use?
"Re-compress images" for scans and photo-heavy documents — it re-encodes pictures as JPEG at the quality and size you choose and keeps text as text. "Lossless" only rewrites the file structure. "Rasterise pages" gives the smallest file but turns every page into an image.
Why did my file barely shrink?
Text-only PDFs and files whose images are already efficient JPEGs cannot be reduced much without losing quality; the tool keeps the original rather than produce a larger file. The Size Analyzer shows what the bytes are spent on.
Is "optimise PDF" the same as compress?
For file size, yes — this is the page. Optimising for Fast Web View (linearisation) and for printing (N-up, booklets) have their own tools.
Is my PDF uploaded to a server?
No. The file is opened and processed by your browser with WebAssembly and JavaScript; nothing is sent to Mutqan Soft or anyone else. Close the tab and the file is gone. Downloading the result needs a free account — no subscription, no payment.
What are the size limits?
Up to 250 MB and 2,000 pages per operation on a desktop browser, 60 MB and 500 pages on a phone, because everything runs in your device’s memory. The tool tells you before it starts if a file is over the limit.
Can I use it on a password-protected PDF?
Yes, if you know the password: the file list asks for it once and every operation then works on the decrypted pages. The output is saved without a password unless you protect it again.
Technical notes
Image mode walks every image XObject, decodes DCT (via the platform decoder) and Flate/raw images (own decoder with PNG/TIFF predictors, Gray/RGB/CMYK/Indexed), downsamples, re-encodes as JPEG and replaces the stream in place — keeping soft masks when dimensions allow. Structural mode copies pages into a fresh file with object streams. The file is only replaced when the result is smaller.
References
- ISO 32000-2:2020 — Document management — Portable document format — Part 2: PDF 2.0The current PDF standard: file structure, objects, pages, fonts, annotations, forms and encryption.
- PDF Reference — ISO 32000-1:2008 (PDF 1.7), Adobe copyThe freely readable PDF 1.7 specification: page trees, content streams, metadata, permissions and the standard security handler.