Image Compressor

How to use the Image Compressor

1. Drop in the image you want to shrink.
2. Either drag the quality slider, or switch to target mode and type the size you need to hit.
3. Add a width limit if the image is larger than it needs to be — this usually saves more than quality alone.

Resize first, then compress

The most common reason a photo is too large is that it is far bigger than it will ever be displayed. A 4000-pixel-wide camera image shown in a 800-pixel column carries twenty-five times more pixels than the screen can use.

Because file size scales with the number of pixels, halving the width quarters the data before compression even begins. That is a much bigger lever than the quality slider, and it costs nothing visually if the image was oversized to start with. Set a width limit that matches where the image will actually appear.

How target-size mode works

There is no formula that maps a quality setting to a file size — it depends entirely on image content. A photo of a blank wall compresses far better than one of a forest at the same setting.

So the tool searches. It encodes at a mid quality, checks the result against your target, and narrows the range from there, converging in about nine attempts. The output is the highest quality that still fits under your limit.

If the target cannot be reached even at minimum quality, the tool says so rather than silently returning something too large. In that case, reduce the dimensions too.

Why compressing an already-compressed file often fails

Recompressing a JPG that was already saved at quality 70 will not help much, and can make things worse. Each lossy pass discards information and adds artefacts, which the next pass then has to encode faithfully — spending bits on the damage rather than the picture.

If the tool reports the result is larger, the source was already efficiently compressed. Reduce the dimensions instead, or switch the output to WebP, which can genuinely beat JPG on the same pixels.

Frequently asked questions