Compress PDF

PDF ToolsPopularPremium

Shrink a PDF losslessly, or downscale and re-encode its images for a much smaller scan.

Nothing Is UploadedRuns In This Browser TabNo Account Needed

How It Works

1

Upload / Input

Select your files or paste your content to get started.

2

Process

The work happens in this browser tab. Your file never leaves your device, so nothing is queued, stored or transmitted.

3

Download / Result

Get your results instantly. Download files or copy output in one click.

Features

Lossless, Balanced or Strong: you choose how much image quality to trade
Balanced and Strong re-encode images at about 150 / 100 dpi
Shows the measured before and after size, not an estimate
Text, links and bookmarks are preserved in every mode
Runs in your browser — the file is never uploaded

How to use Compress PDF

  1. Add the PDF

    Drop in the file. Nothing is uploaded — the re-save runs in your browser.

  2. Choose a mode

    Lossless only repacks the file structure and never touches an image. Balanced and Strong also downscale and re-encode the images inside the PDF (about 150 and 100 dpi), which is where a scanned document’s weight is. An image that cannot be read is kept as it was.

  3. Re-save

    Processing scales with the number of objects in the document rather than with its page count, so a long text document can finish faster than a short image-heavy one.

  4. Read the measured result

    The result screen reports the real before and after sizes. If the difference is negligible, the file’s weight is in its images and no structural re-save can reach it.

When to use Compress PDF

  • Tidying a PDF produced by a generator that wrote a bloated object table
  • Normalising a document that several different editors have passed through in turn
  • Taking the last slice off a file that is only just over a limit
  • Shrinking a scanned PDF (Balanced or Strong) or repacking one without altering a pixel (Lossless)

FAQ

Often very little, and on an already-optimised file sometimes nothing at all. This tool rewrites the document’s internal object table into a more compact form; it does not touch the images, which are what actually make a PDF large. The result screen reports the size it measured rather than a figure promised in advance.
No. Nothing is re-encoded or downsampled. Text, vectors and images come out exactly as they went in — that is the point of a structural re-save, and also why the saving is modest.
No. This tool takes one PDF at a time.
Lossless only repacks the file structure. Balanced downscales images to about 150 dpi at JPEG quality 0.7. Strong goes to about 100 dpi at quality 0.5 for the smallest file. If an image cannot be read it is kept as it was.
Because it was already packed at least as tightly as this method can pack it. The tool never returns more than the lossless re-save, and tells you when there is nothing to gain: keep the original.
In Lossless mode, no: not a single pixel changes. Balanced and Strong downscale images to about 150 and 100 dpi, so use Lossless for anything you must print or submit at full resolution.
Because they re-encode the images inside it at a lower quality. Balanced and Strong do exactly that; Lossless does not, so its saving is far smaller.
Vector text is never touched. Text inside a scanned page is an image, so Strong (about 100 dpi) can soften small print; Balanced keeps it readable and Lossless leaves it exactly as it was.

Related Tools

About Compress PDF

PDF file size is almost always driven by images, not by structure. A hundred pages of plain text is a few hundred kilobytes; a single scanned page can be several megabytes on its own. Lossless mode does not touch the images: it rewrites the document’s internal object table into a more compact form and saves it again. Balanced and Strong also downscale the images to about 150 and 100 dpi and re-encode them as JPEG, which is where a scan loses most of its weight.

In Lossless mode the honest expectation is a small one. A PDF written by a careless producer, or one that several editors have rewritten in turn, carries redundant structure and can lose a noticeable slice. A file already saved by a modern optimising producer has nothing left to give, and the rewrite can even leave it fractionally larger — which is why the result screen shows a measured figure rather than a promise.

If you need a scanned document to genuinely get smaller, the lever is the scan itself: rescan at 300 DPI in greyscale rather than 600 DPI in colour. Balanced and Strong do re-encode the images inside an existing PDF, which is lossy: text stays sharp at 100 to 150 dpi, but fine photographic detail is reduced, so keep the original and compare before you send it.

Tips and limits

  • Lossless gives a small saving because images are untouched. For an image-heavy file choose Balanced or Strong.
  • Text stays selectable and links and bookmarks survive in every mode. Only Lossless keeps images at their exact resolution.
  • If the output is larger than the input, the original was already packed as tightly as this method can pack it — keep the original.
  • Running it twice achieves nothing. The second pass is handed the same structure the first one produced.
  • For a real reduction on a scan, use Strong, or rescan at 300 DPI greyscale in the first place.