Compress PDF to 500 KB

Job portals, university applications and government sites often reject PDFs over 500 KB. This tool works towards a size you set rather than a vague "compression level": it removes waste without touching quality first, then recompresses photos and scans just enough, and only as a last resort turns pages into images. It aims for 500,000 bytes, so the file passes whether a site counts 1 KB as 1,000 or 1,024 bytes. Everything runs in your browser.

How to use this tool

1. Choose the PDF (or several — each is compressed separately).
2. Keep 500 KB or type another target, such as 300 KB or 2 MB.
3. Optionally tick greyscale for scans, or untick "turn pages into images" to keep text selectable no matter what.
4. Download the result and check "Target reached" and the method used.

Formula

Stage 1 is lossless: unused objects are removed and uncompressed streams are compressed. Stage 2 re-encodes photo images as JPEG, trying settings from 150 dpi at 80% quality down to 30 dpi at 20%; a binary search finds the best-looking setting whose predicted size fits, and the real file is then saved and checked. Stage 3 renders each page to a JPEG and searches the same scale using sample pages.

Resolution is measured against the largest page side, so a 150 dpi limit on an A4 page keeps images up to 1,754 pixels on their long edge. Images that would get bigger are left exactly as they were.

Example

A 6-page scanned application of 3.8 MB is 6 full-page JPEGs. Lossless clean-up saves little, so stage 2 tries settings until it finds one that fits — for example 96 dpi at 50% — giving about 460 KB with the text layer, if any, still selectable. A text-only PDF of 900 KB with embedded fonts may need stage 3 instead.