Image Tools

Image tools are where running in the browser matters most. Most online image services upload your file, keep it for a while, and often watermark the result or cap the size. Everything in this category is processed on your own device with the browser's Canvas and WebAssembly builds of the same libraries the desktop programs use — MozJPEG, libwebp, libavif and OxiPNG for encoding, libheif for iPhone photos, Tesseract for text recognition — so a passport photo or a client's design never leaves your computer.

Compress is the tool most people come for: it shows the before and after side by side and keeps the original if the result would be larger. Convert moves between JPG, PNG, WebP and AVIF; HEIC to JPG handles the format iPhones shoot in. Resize, crop and rotate do what their names say, and watermark stamps text or a logo across a batch.

Two tools run a model rather than an encoder. Remove background uses a segmentation network (ISNet) running on your graphics card through WebGPU where the browser has it, falling back to the CPU where it does not. Image to text reads printed text in twelve languages, including Chinese, Japanese and Korean; handwriting is outside what it can do. Make a QR code and read a QR code round out the set — the reader works from a screenshot or a photo of a poster.