Demos

UchenML models you can poke at, right in the browser.

Every demo on this page runs local inference in WebAssembly — the same C++ model definition, parameters, and inference path that ships in a native binary. No backend, no hosted inference API.