About AiPal

AiPal is a collection of free utilities that run small open-source AI models entirely inside your browser. When you use a tool, the model is downloaded once — from our own servers where available, otherwise from the model's public repository — and cached locally. After that, everything — including your files — stays on your device.

Inference runs on WebGPU where available, with a WebAssembly fallback for older browsers and most mobile devices. Heavy processing happens in Web Workers, so the page stays responsive.

The project is built with Next.js, TypeScript and Transformers.js. It is designed to be extended: every tool is a configuration entry plus a processor, and tools that use the same kind of model share one cached download.