Image compressor

How target-size compression works

Most compressors give you a quality slider and leave you to guess. This one works backwards from the limit the form gave you:

1. Decode once. The file is read at full resolution in your browser. Phone photos are rotated upright using their EXIF orientation before anything else happens.
2. Apply your long-edge limit. If you chose 480 px, the longest side is resized to 480 px. Downscaling in steps keeps edges clean.
3. Search for quality. The image is encoded several times between 28% and 95% quality. The tool keeps the highest setting whose file is still at or below your target.
4. Shrink only if needed. If even 28% quality is too large, the long edge is reduced by about 18% and the search repeats. It stops at 96 px, which is too small for most uses, and tells you instead of pretending the target was met.

What survives and what is removed

The new file contains only pixels: no EXIF camera data, no GPS location, no thumbnails and no editing history. That is good for privacy, but it also means the date the photo was taken is lost. If a form asks for the original capture date, keep a copy of the original file.

JPG cannot store transparency. A logo or signature with a transparent background is placed on white before encoding. Choose WebP if transparency matters, or use the image converter to keep PNG.

Getting the best result at a small size

• Crop first. A tight crop of the face or signature spends the available bytes on what matters.
• Use a plain background. Texture, shadows and gradients cost bytes; a uniform background compresses almost for free.
• Do not upscale. A small source enlarged to 480 px looks soft and compresses badly. Upload the largest original you have and let the tool shrink it.
• Check the form's wording. “20 KB” may mean at most 20 KB, or approximately 20–50 KB. Stay at or below the hard limit.

For the full method, format differences and country-by-country form limits, read the image tools guide.