Skip to the tool

English Deutsch Español

On this device

Organize12

Merge PDF Split PDF Extract pages Split PDF by size Delete pages Organize PDF Rotate PDF Crop PDF Pages per sheet Add page numbers Add watermark Stamp PDF

Optimize7

Compress PDF Repair PDF Optimize PDF for web Flatten PDF Edit PDF metadata Remove PDF metadata Edit PDF bookmarks

Convert15

PDF to JPG PDF to PNG PDF to WebP Extract images from PDF PDF to text PDF to Markdown OCR PDF JPG to PDF PNG to PDF WebP to PDF HEIC to PDF TIFF to PDF SVG to PDF Text to PDF Markdown to PDF

Edit and sign5

Edit PDF Fill PDF form Sign PDF Digitally sign PDF Compare PDF

Secure3

Protect PDF Unlock PDF Redact PDF

All tools
Compress PDF

Compress PDF

Make a PDF smaller and see exactly which bytes go and what is lost.

Drop a PDF here

Pages appear straight away — nothing is sent anywhere.

Read on this device. No account, no upload. Drop as many files as you like — every one of them is processed.

Related tools

Optimize PDF for web
Split PDF by size
Repair PDF

Settings

Nothing starts until you press the button, and nothing is sent when it does.

Level

Keep image qualityFull resolutionRe-encodes images without lowering resolution. Safe for print work.BalancedRecommended150 dpiImages are re-encoded and downsampled to 150 dpi. Prints cleanly on office paper.Smallest file96 dpiFor email and web. Photographs soften noticeably at full zoom.

Estimated result

Pick a file and every level shows the size it would actually produce.

Advanced

Images

Re-encode imagesLoses detailPhotos and scans are compressed again with a smaller JPEG. Text and vector drawings are untouched.Drop colour from imagesLoses detailRight for scanned paperwork, wrong for photographs. Colour cannot be brought back.

Image resolutionLoses detail

Fewer dots per inch means fewer pixels to store. 150 dpi prints cleanly; 96 dpi is for reading on screen.

Image qualityLoses detail

How much detail each image keeps. Lower quality means a smaller file and softer photographs.

Structure

Pack internal objectsNothing lostStores the document's object graph in compressed streams. The largest saving that costs nothing.Drop unreferenced objectsNothing lostRemoves objects no page points at, usually leftovers from an earlier edit.Store repeated objects onceNothing lostKeeps a single copy of an object that appears more than once.Optimise for the webNothing lostReorders the file so a browser can show page one before the rest arrives. Adds about one percent.

What to remove

Remove metadataLoses detailTitle, author, producer and edit timestamps, plus the XMP packet behind them.Remove attachmentsLoses detailDeletes files embedded inside the document. They cannot be recovered from the output.Remove comments and highlightsLoses detailDeletes notes, highlights and stamps. Form fields are a separate structure and stay.Remove stored page previewsNothing lostReaders draw their own previews, so removing the stored ones changes nothing on screen.Remove embedded scriptsNothing lostDeletes JavaScript carried in the document. Page content is untouched.Remove the tagging treeLoses detailScreen readers use these tags to read the document in order. A smaller file bought with the document's accessibility.

Kept at every level

Selectable text, so the document stays searchable.
Bookmarks, internal links and page order.
Form fields, unless you remove them above.
The file you started with — pdfhq works on a copy.

Read more

How to reduce PDF file size
What DPI and quality change
PDF privacy on the web

Add a file to start

Runs in this tab. Your file is not uploaded.

Tools Task Recent Settings