Compress PDF rebuilds a PDF to reduce its file size without sending the document anywhere. Use lossless clean-up to rebuild a text document and clear its metadata, or strong compression to rebuild scans and photo-heavy files as JPEG pages. The original is never changed.
How to compress a PDF
- Drop your PDF or click Choose file.
- Choose Clean up (lossless) for text documents, or Strong compression for scans and photo-heavy files.
- For strong compression, choose a resolution, set the JPEG quality and tick Convert pages to greyscale if colour is not needed.
- Click the button and wait. Strong compression shows page-by-page progress; long documents take longer and use more memory.
- Download
<name>-compressed.pdf. The file on your device is left untouched.
Clean up versus strong compression
Clean up rebuilds the PDF with a more compact internal layout and removes document properties (title, author, subject, keywords, creator, producer and dates) together with the XMP metadata block. Pages are copied rather than redrawn, so text stays selectable, images keep their original quality, and links, bookmarks and form fields remain. Savings are usually small — a few per cent, sometimes none, and occasionally the file is slightly larger. Rebuilding also invalidates digital signatures, so keep the signed original.
Strong compression draws each page at the chosen resolution, encodes it as a JPEG and builds a new PDF from those images. Page size and orientation are preserved. It can shrink a wasteful scan substantially, but it loses selectable text, links, interactive form fields and bookmarks, and invalidates digital signatures.
| Clean up (lossless) | Strong compression | |
|---|---|---|
| Text | Stays real text | Becomes part of an image |
| Best for | Contracts, reports, forms | Scans, photo-heavy brochures |
| Links and forms | Kept | Lost |
| Typical result | Small saving, sometimes none | Can be large, sometimes larger |
What strong compression does to text and search
Every page becomes a picture. You can read, print and zoom it, but the words are no longer text. Find, select and copy stop working, screen readers have nothing to read, links are not clickable and form fields are not fillable.
If you need a searchable file, use clean-up instead. If you need both a small file and searchable text, run OCR afterwards to add a text layer; this tool does not perform OCR. Strong mode is a poor trade for text-heavy documents, where it often makes the PDF larger as well as less useful.
Choosing DPI and JPEG quality
DPI controls how many pixels describe each page. The PDF's physical page size stays the same.
| Resolution | A4 page in pixels | Good for |
|---|---|---|
| 72 DPI | 595 × 842 | Quick previews and very small files |
| 110 DPI | 909 × 1286 | Reading on screen, email attachments |
| 150 DPI | 1240 × 1754 | The best default for most scans |
| 200 DPI | 1654 × 2339 | Small print and documents you may print |
JPEG quality runs from 40 to 90, with 70 as the default. Values around 40 to 55 compress aggressively and can leave visible blocks around small text; 60 to 75 suits most scans; 80 to 90 keeps more detail for photographs, at a larger size. Resolution affects legibility more than quality: small text rendered at 72 DPI stays blurry even at quality 90.
Greyscale can reduce colour scans that contain little meaningful colour, such as typed forms. It removes colour permanently.
Scanned PDFs and text PDFs
A scanned PDF usually contains one image per page. Strong compression suits these files, especially when the scan is oversized or uses inefficient compression. Lower resolution and quality, and greyscale for black text, often help. If the scan is already a 150 DPI JPEG at moderate quality, re-encoding it can add softness without shrinking much.
A text PDF is made of characters, fonts and vector shapes. Strong compression usually makes it larger and always removes search and selection, so use clean-up. For a mixed document, split the file, compress only the scanned pages, then join the parts.
Long scans need memory. At 200 DPI, an A4 page is about 3.9 million pixels; 100 pages is roughly 390 million before JPEG compression. Each canvas is released after encoding, but the page images stay in memory until the PDF is saved.
Why some files get bigger
Most PDFs are already optimised, with subsetted fonts, compressed images and often object streams. Clean-up mostly removes metadata and rebuilds the structure, so the saving is limited to what those parts occupy. Metadata is often only a few kilobytes, negligible in a 5 MB file.
Strong compression can easily grow a text PDF. A typed page might take 50 KB as text and fonts but several hundred kilobytes as a 150 DPI JPEG. JPEG is lossy, so re-encoding an efficient original can increase its size as well as soften it.
If the result is not smaller, keep the original. Try lower settings in strong mode, or clean-up for a text PDF.
Privacy
PDF.js renders the pages and pdf-lib writes the new file inside your browser, so files never leave the browser. Nothing is uploaded or copied to an account. The PDF.js worker is served from this site but runs locally, and closing the tab discards the document. It suits private documents on a device you trust.
Related PDF tools
- To turn pages into images and inspect them, use PDF to JPG.
- To build a PDF from images, use JPG to PDF.
- To split a long document before compressing scanned pages, use Split PDF.
- To trim white space or scanner edges, use Crop PDF.
- To create a scan from photos first, use Scan to PDF.
Compress PDF FAQ
Is this PDF compressor free?
Yes. There is no sign-up, no watermark and no page limit; very large files depend on your device's memory.
Are my PDFs uploaded to a server?
No. The file is read from your device and processed entirely in the browser. Nothing is sent to Lembog or any third party.
Will strong compression make the text unreadable?
It can stay readable as an image, but the text is no longer selectable or searchable. Legibility depends mainly on resolution: choose 150 or 200 DPI for body text.
Why is my compressed PDF larger than the original?
This is common with text-heavy PDFs in strong mode and with files that were already optimised. Rebuilding pages as JPEGs adds image data, and re-encoding can increase size. Try lower settings or keep the original.
Which DPI should I choose?
Start with 150 DPI, a good balance for most scans and screen reading. Use 72 or 110 DPI for the smallest files, and 200 DPI for small print or printing.
Can I compress a scanned PDF?
Yes, and strong compression is usually the better mode. Greyscale can help with colour scans of black text. If the scan is already small and efficiently compressed, the saving may be modest, and repeated JPEG compression can soften fine text.
Can I compress a password-protected PDF?
No. Remove the password first in a PDF reader by saving an unprotected copy of a document you own or have permission to modify, then compress the copy.
Does clean-up remove all metadata?
It removes the usual document properties and the XMP metadata block, but it does not alter page content. Comments, attachments, hidden layers and other embedded data remain, so clean-up is not a redaction or sanitising tool. Strong compression leaves out hidden content, but it is not a substitute for proper redaction.