pdfpix

Compress PDF to 100 KB

or drop one here
Compression level
Target size
Choose a PDF file first.

100 KB is 102,400 bytes, and it is a hard ask. The public bench met a 100 KB target on 23 of 56 cases, 41.1%, the lowest reach of the four targets it ran. Scans usually make it. Text documents with embedded fonts usually do not, because the fonts alone can outweigh the cap.

How to compress a PDF to 100 KB

  1. Choose the PDF

    Press Choose a PDF, or drop the file on the page. The file is read in this tab and is not sent anywhere.

  2. The target is already set

    This page opens at 100 KB, so there is nothing to type. Pick a different level, or enter your own smaller number in KB, if you want another size.

  3. Press Compress

    Four methods race: copy the pages into a fresh file, re-encode the placed images, cut each embedded font down to the glyphs the pages draw, and, only when no other method reached the target, redraw the pages as images with the text kept on top.

  4. Read the check, then download

    Every candidate is opened again and its characters are counted against the source. The page shows the size, the method that won and the characters kept before it offers the download.

The four methods are described in full, with the measurements behind them, in how it works.

What fits under 100 KB and what does not

A one or two page scan fits. A photo of a document fits. A signed form that started life as a Word export usually fits once its fonts are subset. What does not fit is a long report with charts, or any file where five or six embedded font programs carry most of the bytes.

An embedded font ships with every glyph it was designed with, while the document draws a few dozen. The compress tool cuts each font down to the glyphs actually used, which is often the single biggest saving on a text file, and still not enough at 100 KB.

The measured reach at this target

100 KB was one of the four bench targets, so this number is measured: 23 of 56 cases met it. That is the hardest of the four, and it is hard because the ask is large. The median source in that group was several times the cap, so meeting it means cutting most of the file away.

When the engine cannot reach it, it ships the smallest file it proved and reports the size. Over the whole run the median cut was 59.2% and the ninetieth percentile was 88.1%, so a miss is still usually a much smaller file than the one you started with.

Before you try 100 KB at all

Check the form again. Many upload fields that say 100 KB mean 100 KB per page, or accept a larger file with a different extension. It is worth thirty seconds, because a 100 KB target on a ten page scan gives you 10 KB of pixels per page, and text on that page stops being readable.

If the cap has any give in it, a 150 KB target or a 200 KB target keeps a lot more image detail for the same work. The compression guide explains which lever moves first and why.

Questions

Can any PDF be compressed to 100 KB?
No. A file with photographs on every page has a floor, and on a long document that floor sits well above 100 KB. The bench met a 100 KB target on 23 of 56 cases. The rest came back smaller but still over the cap, with the reached size reported.
Why do fonts make a 100 KB target so hard?
An embedded font carries every glyph it was designed with, not the ones the page draws. Several fonts in one file can add up to more than the whole 102,400 byte budget before a single page of content is counted.
Does a 100 KB target make the scan unreadable?
On one or two pages, usually not. On ten pages it can. A 100 KB target spread over ten scanned pages leaves about 10 KB of image per page, which is below what small print survives. Fewer pages or a larger cap is the fix.
Is the file uploaded anywhere to reach 100 KB?
No. The compression runs inside the browser tab. The PDF is read from disk into memory, worked on there, and written back to your downloads folder. Nothing is sent to a server and nothing is stored.

The other pdfpix tools

All 30 of them run in a browser tab like this one. Each reads its own output back and checks it before offering the download, and none of them needs an account.

Where to go next