pdfviz

Compress PDF
to 250 KB

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

250 KB is 256,000 bytes at 1,024 bytes to the kilobyte, which is how the tool counts. That number is the reason the cap exists: forms sized at a quarter of a megabyte clear against it. The bench ran 250 KB on a two-page scan and landed at 300.8 KB from 498.4 KB in all three runs: a miss, and proof that two pages do not guarantee 250 KB.

How to compress a PDF to 250 KB online

  1. Choose the PDF

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

  2. The target is already set

    This page opens with the target fixed at 250 KB, so there is nothing to type. Pick a different compression level if you want, or enter your own number in KB for a different size.

  3. Press Compress

    Four methods race against each other: 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. The whole process runs in your browser through WebAssembly, which means your file is never uploaded and no account is needed.

Why 250 KB is really 256,000 bytes

A 250 KB limit usually started life as a quarter of a megabyte. Counted in binary all the way that is 262,144 bytes; the tool counts 250 KB as 256,000 bytes at 1,024 bytes to the kilobyte. The two readings differ by about 2%, so aiming slightly under the published cap is the safe path.

The compress tool uses 1,024 bytes to the kilobyte, so a 250 KB target here is 256,000 bytes and it stops below that. That reading matches how the caps are usually implemented, and it clears them. If a form still refuses the file, it is doing plain decimal arithmetic, and 240 KB clears both.

This is the kind of detail that matters on exam boards, visa applications, and insurance portals. They publish a round number and enforce it with a specific byte count, and the two do not always agree. Aiming slightly under the published cap is the safe path.

What fits under a 250 KB cap

One or two scanned pages at a quality where printed text stays sharp. A text document of any reasonable length, once its fonts are subset. What does not fit is a colour photograph at full resolution, or a scan running past six or seven pages.

No hit rate is measured for 250 KB itself. Reach rises with the cap, and the kind of document matters more than the exact number. A text document almost always clears 250 KB through font subsetting alone, while a scan depends on the page count and the image content.

For a one or two page scan, the image re-encoding method picks a quality that keeps printed text readable, but it can still miss 250 KB: the bench two-page scan stopped at 300.8 KB. For a longer scan, each page gets a smaller share of the budget and small print starts to break up.

What happens when compression misses 250 KB

You get the smallest file the engine could prove, together with the size it reached and the method that won. That is the information you need: a miss at 270 KB means one page or one quality step away, and a miss at 800 KB means the document has to change.

Any candidate that keeps less than 98% of the source's non-space characters is refused. If the best candidate falls short of the target, it ships the smallest one that passed verification and tells you the real floor. That number is the honest answer for your document.

A 200 KB target is the next step down if the form turned out to be stricter, and a 300 KB target is the next step up if it turned out to be looser. Both use the same engine and the same checks.

Real use cases for a 250 KB target

Exam and visa upload forms that publish a quarter megabyte cap. The tool counts the same way the cap was originally sized, so the file clears it without a second attempt.

Insurance claim attachments where a one or two page scan needs to fit under a strict limit. The image method keeps printed text sharp at this budget.

Job application portals that ask for a CV or cover letter under 250 KB. A text document usually clears it through font subsetting alone, with no pixel touched.

Questions about compressing a PDF to 250 KB

Is a 250 KB limit 250,000 or 256,000 bytes?
Either, depending on the form. The tool counts 256,000 bytes at 1,024 bytes to the kilobyte. If a form still refuses the file, it is doing plain decimal arithmetic, and 240 KB clears both.
How many scanned pages fit in 250 KB?
Not necessarily two. The bench ran a two-page scan at 250 KB and landed at 300.8 KB from 498.4 KB in all three runs. Past two pages the per page budget falls further and small print starts to break up. Splitting the scan is better than pushing the quality lower.
Does the 250 KB target work on a phone?
Yes. Files this size are well inside what a phone browser handles, and the compression runs in the tab exactly as it does on a desktop. The file is read in this browser tab and is not sent to any server.
What if I need 250 KB and the file is a text document?
Text documents are the easier case here, because their weight is fonts and structure rather than pixels. Font subsetting is usually the winning method, and if it still misses, removing pages is the only lever left.
Will the text stay selectable after compression to 250 KB?
Yes. Every candidate is re-opened and its characters are read back and compared with the source. Anything below 98% retention is refused.
Can I type a custom target instead of 250 KB?
Yes. The target size row has a text field where you can enter any number in KB or MB. Pick a chip for a common size, or type your own for a form with a non standard cap.

The other pdfviz tools

They all 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