This project implements a state-of-the-art image classification model using Vision Transformers (ViT). It achieves top-1 accuracy on the CIFAR-10 dataset.