Compress a PDF (scans first)
DocumentsEverything runs locally in your browser — nothing is uploaded
Drop files here, or click to choose
PDF · up to 100 MB per file
Make a PDF smaller, and here is the trade-off up front: this tool re-renders every page as a JPEG and rebuilds the file, so text becomes an image — not selectable, not searchable, pixelated when zoomed. For scans (where every page already is an image) that is pure gain; for digitally typeset PDFs (text and vectors) it can come out larger, and the tool tells you so rather than handing back a bigger file. The three settings: resolution (halving the DPI quarters the pixels — the biggest lever), quality (70% is usually indistinguishable), and greyscale (halves black-and-white documents again). If keeping selectable text matters, do not use this tool — that needs the lossless route of re-encoding only the embedded images, which is unreliable in a browser (replacing image streams means editing the object table, and getting it wrong yields a file that will not open).
Features
- Re-renders each page as JPEG; scans often shrink by 70–90%
- Resolution of 150, 120, 100 or 72 DPI
- Three JPEG quality steps
- Optional greyscale — halves black-and-white documents again
- Compress selected pages only (for a document that is mostly text with a few photo pages)
- The physical page size never changes — compressing should not change print size
- Says so when the result is larger, and suggests keeping the original
- Nothing is uploaded and no server is involved
How to use
- Drop a PDF in (scans work best)
- Choose the resolution and quality; tick greyscale for black-and-white documents
- Download and flip through a few pages to check legibility
- If it is a text PDF that will not shrink, keep the original
FAQ
- Why can I no longer select the text after compressing?
- Because the tool re-rendered every page as an image. The text is now pixels with no text layer — that is the price of the strongest compression. Keeping selectable text means a different route (re-encoding only the large embedded images while leaving text and vectors alone), which requires replacing image streams in the PDF's object table — pdf-lib's public API cannot do that, and getting it wrong produces a file that will not open, which is worse than a file that did not shrink. So this tool offers the one reliable path and states its cost plainly.
- It came out larger. Why?
- Because the document is mostly text and vector art, not scans. Text costs tens of kilobytes in a PDF (it is instructions); rebuilt as an image it becomes hundreds of kilobytes of pixels. The right move is to keep the original — that file is already small (text PDFs are typically a few hundred kilobytes, with nothing to compress). If the goal is a smaller file for a phone, try 72 DPI, accepting visibly soft text.
- Can I hit a specific size, like under 5 MB?
- Not directly — the size depends on page count and content, so you approach it by adjusting DPI and quality. As a rule of thumb, one A4 scan at 120 DPI and 70% quality is about 100–200 KB, so 40 pages lands around 4–8 MB. To go smaller, drop to 100 DPI or turn on greyscale (half the size again for black-and-white content). When an upload limit is the problem, greyscale alone often settles it.
- Is the file uploaded?
- No. pdf.js renders, canvas encodes, and pdf-lib reassembles, all in browser memory with no network requests, and it works offline.
Related tools
PDF to Image
DocumentsRender each page of a PDF to an image at 96, 150 or 300 DPI. Zero-padded names keep page order, and you can export selected pages only.
Image Compressor
ImagesCompress an image to a target size (say under 200 KB): the tool binary-searches for the highest quality that fits. Quality steps and size limits too.
PDF Info
DocumentsSee a PDF's page count, each page's size and orientation, whether it is encrypted, its form fields and its metadata. Exports as text or JSON.
PDF Merge
DocumentsCombine several PDFs in the order you choose, or take only certain pages from each. Rename the result; contracts never leave your browser.
PDF Split
DocumentsExtract pages into one file, or split into a file per N pages. Output names carry the page range, so a file sent on still makes sense. Nothing is uploaded.
Images to PDF
DocumentsCombine JPEG and PNG images into a single PDF, one image per page, on pages that match each image or on A4 and other fixed paper sizes.