How is this different from Compress PDF?

Compress PDF renders each page as an image and rebuilds the file — very small, but the text becomes part of the picture. This is lossless structural optimization: page content doesn’t change and text can still be selected, searched and copied, which suits contracts, papers and e-books. Pick whichever fits.

Why did my file barely shrink?

Files produced by professional software are usually already compressed and free of duplicates, so there’s little structural fat; most of their size is photos. Turn on image downsampling or use Compress PDF instead. If the optimized file isn’t smaller, you simply get the original back.

Will downsampling make images blurry?

Only images whose resolution, at the size they’re actually shown on the page, is clearly above the target DPI are processed — say a 4000-pixel photo shown as a small thumbnail. 150 DPI suits screen reading; use 200–300 DPI for print. JPEGs are re-encoded at your chosen quality; lossless images only lose resolution and stay lossless.

Is my file uploaded? Are forms or bookmarks removed?

Nothing is uploaded; it all happens in your browser. Forms, bookmarks, links and annotations are kept — only the metadata, thumbnails, attachments and scripts you tick are removed. Without scripts, form fields that calculate automatically stop doing so.

🏎️

🏎️Optimize PDF - Lossless Size Reduction, Selectable Text

Unlike Compress PDF, which turns every page into an image, this tool slims down a PDF’s structure without changing how its pages look: text, vector graphics and layout stay exactly as they were, and text remains selectable and searchable. PDFs are often big because the same image or font is embedded over and over (common after merging files or exporting page by page), because content streams aren’t compressed or use ASCII encoding, or because they carry thumbnails, private data from editing apps and old objects nothing refers to any more. Optimization merges identical font programs, images, ICC profiles and font dictionaries by content fingerprint; switches uncompressed, ASCII- and LZW-encoded streams to Flate and re-compresses existing Flate streams, keeping the result only when it’s smaller; optionally removes document info and XMP metadata, page thumbnails and PieceInfo, attachments and JavaScript; then removes unused objects and packs the rest into object streams. For extra savings, turn on downsampling of only those images shown at too high a resolution: the DPI is calculated from each image’s actual size on the page, only images clearly above the target are processed, JPEGs stay JPEG, lossless images stay lossless, and every other image is left byte-for-byte untouched. A before / after breakdown by page content, images, fonts, metadata, thumbnails, attachments and file structure shows exactly where the space goes. Everything runs in your browser; files are never uploaded.

Free · No sign-up · Runs in your browserLast updated: 2026-10-26

Loading tool...

How to use Optimize PDF

1Upload a PDF and see how its size breaks down into content, images, fonts and more
2Tick the lossless options and the data to remove; optionally enable image downsampling and set the target DPI
3Click Optimize & download and compare the categories before and after

FAQ

How is this different from Compress PDF?: Compress PDF renders each page as an image and rebuilds the file — very small, but the text becomes part of the picture. This is lossless structural optimization: page content doesn’t change and text can still be selected, searched and copied, which suits contracts, papers and e-books. Pick whichever fits.
Why did my file barely shrink?: Files produced by professional software are usually already compressed and free of duplicates, so there’s little structural fat; most of their size is photos. Turn on image downsampling or use Compress PDF instead. If the optimized file isn’t smaller, you simply get the original back.
Will downsampling make images blurry?: Only images whose resolution, at the size they’re actually shown on the page, is clearly above the target DPI are processed — say a 4000-pixel photo shown as a small thumbnail. 150 DPI suits screen reading; use 200–300 DPI for print. JPEGs are re-encoded at your chosen quality; lossless images only lose resolution and stay lossless.
Is my file uploaded? Are forms or bookmarks removed?: Nothing is uploaded; it all happens in your browser. Forms, bookmarks, links and annotations are kept — only the metadata, thumbnails, attachments and scripts you tick are removed. Without scripts, form fields that calculate automatically stop doing so.