Skip to content

Image Compressor

Shrink JPG and WebP file size with a quality slider and live before/after preview.

Runs in your browser

Drop a JPG or WebP image, drag the quality slider, and watch the file-size readout drop. A side-by-side preview shows the visual cost of the compression. Both the original and the compressed copy stay in your browser - nothing is uploaded. PNG is intentionally not supported here because it's lossless; convert PNG to WebP via the image-converter tool first if you want to compress it.

Drop a JPG or WebP here or click to choose
JPG or WebP - PNG won't compress (use image-converter first)

How to use it

  1. Drop a JPG or WebP

    Drag and drop or click to pick. PNGs are not supported here - convert first via image-converter.

  2. Drag the quality slider

    Watch the file-size readout drop and the side-by-side preview change. 75-85 is the usual sweet spot.

  3. Download the compressed copy

    Filename gets `-compressed` appended. The original is untouched.

What is it?

An image compressor reduces the file size of a JPG or WebP image by re-encoding the pixels at lower quality. The browser's Canvas API handles the re-encoding; you set the quality on a 0-100 scale and the encoder discards visual detail proportionally. The result is a smaller file that looks almost identical at typical viewing sizes - the workhorse trick behind every fast-loading website.

When to use it

Preparing photos for a website before they slow down Core Web Vitals. Shrinking an iPhone photo from 5 MB to 200 KB before emailing it. Compressing user uploads on the way to a storage bucket where bandwidth costs add up. Anywhere file size matters and a 5-10 % visual quality drop is invisible at the viewing size.

Common mistakes

Compressing the same image multiple times - lossy formats lose a little quality each round (generational loss); always compress from the original. Picking 100 % quality and being disappointed by tiny savings - the sweet spot is 75-85 %. And compressing a UI screenshot to a lossy format - the compression banding around sharp edges makes the screenshot look broken.

FAQ

Why is there no PNG option?
PNG is a lossless format - the only way to compress it is to reduce the colour palette (8-bit PNG) or convert to a lossy format. We don't try to fake compression by re-encoding PNG; instead, send the file through image-converter (PNG → WebP or JPG) and then compress from there.
What quality is 'safe'?
75-85% is the standard 'visually lossless' band for photos. Below 70% you start to see blocky compression artefacts on smooth gradients (sky, skin). Above 90% the file barely shrinks. For UI screenshots, stay above 90% or use a lossless format instead.
Will it strip EXIF metadata?
Yes. Canvas re-encodes from raw pixels, so EXIF (camera, GPS, timestamps), the original ICC colour profile, and any embedded thumbnails are stripped. That's usually what you want when publishing to the web; if you need EXIF preserved, use a dedicated photo tool.

More in this category