Goal
Study the components of text-to-image diffusion and how they fit together in a simplified implementation inspired by Imagen.
Current scope
This is a learning project in progress. The repository records the implementation work; it is not a completed reproduction or a claim to the original Imagen model’s results.