We train and fine-tune diffusion models. We also configure large language models to understand user input and pixel data.
