Image Classification

Identify what's in an image using a Vision Transformer (ViT) model that runs entirely in your browser. The model (~90 MB) downloads once on first use and is cached; nothing is uploaded.