Compress PDF
Upload your files here
Compressing a PDF here means reducing its file size by re-encoding its images at a chosen JPEG quality, stripping out metadata, and rewriting the document as a single clean revision. Text is never re-rendered, image pixel dimensions are preserved, and images that re-encoding could corrupt are left untouched, so the savings depend on how much of the file is made of compressible image data.
At a Glance
1 Credit
Per Compress
Image Re-encode
Method
25 MB
Free File Limit
500 MB
Pro File Limit
Key Facts
LittlePDF Compress PDF re-encodes JPEG and plain raster images at one of three JPEG quality levels (0.4, 0.65, or 0.85), strips metadata, and rewrites the file as a single revision; images with transparency or unusual encodings are kept unchanged.
The compression runs entirely client-side in the browser using pdf-lib and the canvas API, so the file is never uploaded to a server.
Each compression costs 1 credit; free accounts get 500 credits per day with a 25 MB file limit, and Pro supports files up to 500 MB.
How to Compress PDF in 3 Steps
Drop the PDF you want to slim down into the upload area, or click to browse.
Pick a compression level: Maximum for the smallest file, Minimum for the best image quality.
Click Compress and download the result; the completion status reports the old and new size and how many images were recompressed.
Why Compress PDF with LittlePDF?
Compression runs in your browser with pdf-lib and the canvas API, so the document is never uploaded.
Three quality levels let you choose the size-versus-sharpness tradeoff instead of accepting one default.
Images that re-encoding could corrupt, such as those with transparency or unusual encodings, are detected and kept as originals.
One credit per file against 500 free credits a day, so compressing documents is effectively free for regular use.
Features
Adjustable Image Recompression
JPEG images and plain raster images are re-encoded at the JPEG quality your level sets: Maximum, Recommended, or Minimum compression. Pixel dimensions are kept; only the encoding quality changes.
Metadata Removal
Author names, creation and modification timestamps, producer and creator strings, and other hidden tags are stripped out. Useful both for size and for not leaking authoring details when you share a file.
Safe-Skip for Fragile Images
Images with transparency, stencil masks, custom decode ranges, or predictor-compressed data are kept as originals rather than risked through re-encoding, and a re-encoded image is only used when it actually comes out smaller. The completion message reports how many images were recompressed and whether any failed and were kept.
Frequently Asked Questions
Last updated: June 2026