Quality-aware compression
Auto quality searches for the smallest result that still meets a perceptual SSIM target instead of relying on a guessed slider value.
Small files, without the mush
Comprimage pairs modern codecs with careful resizing so you can reduce image weight without giving up the detail that matters.
Every decode, resize, comparison, and encode happens on this device. Your images are never sent anywhere.
[ a three-step local pipeline ]
The order matters: decode faithfully, resize without a destructive jump, then encode for the chosen destination.
1. Decode
Your browser opens the source into raw pixels. No server or upload is involved.
2. Resize carefully
Large reductions happen in progressive steps to protect edges and fine texture.
3. Encode locally
A WebAssembly codec runs in a background worker so the interface stays responsive.
[ why the result holds up ]
Professional codecs
MozJPEG, libwebp, AVIF, and JPEG XL provide modern output formats and fine control over effort, subsampling, and lossless modes.
A stronger privacy model
The app never receives your image data because there is no upload endpoint. Every decode, resize, and encode runs entirely in your browser.
[ formats and control ]
Formats and control
Export JPG, PNG, WebP, AVIF, or JPEG XL. Lossy formats expose quality and effort controls; supported codecs also offer lossless output and chroma subsampling choices.