A photo straight off a modern phone camera can easily run 4 to 8 MB. That's no problem for local storage, but it's a real problem the moment you need to email it, upload it to a form with a size limit, or use it on a website where every extra kilobyte slows the page down. The good news: you can usually cut a photo's file size by 70–90% with no visible difference to the eye.
Why images compress so well
Photos are compressed using formats like JPEG and WebP, which throw away detail the human eye is bad at noticing — subtle color gradients, high-frequency texture — while keeping the edges and contrast that your eye actually uses to judge sharpness. That's why a photo can lose 80% of its file size while still looking essentially identical at normal viewing size.
Step by step
- Open the Image Compressor.
- Upload your photo. Nothing is sent to a server — the compression happens right in your browser.
- If you have a specific limit (a form that caps uploads at 100 KB, for example), type that number into the target size field. The tool will search for the quality level that hits it.
- Otherwise, drag the quality slider. Start around 75–80% — that's usually the sweet spot where file size drops a lot but quality loss is invisible.
- Compare the before and after size, then download.
When to use JPEG vs WebP
JPEG is the safe, universal choice — every device and platform accepts it. WebP typically compresses 20–30% smaller than JPEG at the same visual quality, and is well supported by modern browsers, so it's a good choice for websites specifically. If you're not sure, JPEG is the safer default for compatibility.
A common mistake
Resizing the dimensions of an image (fewer pixels) and compressing it (same pixels, less data per pixel) are different operations that both reduce file size, and people often only do one. If an image is both very large in pixel dimensions and needs to be small in file size — say, a 4000px-wide photo that needs to end up under 100 KB — resizing it down first with the Image Resizer before compressing will give you a much sharper result than compression alone, since you're not asking a handful of KB to describe millions of pixels.