The original Stable Diffusion V1 changed the nature of open source AI models and spawned hundreds of other models and innovations around the world. Version 2.0 of Stable Diffusion includes robust text-to-image conversion models trained with a new text encoder (OpenCLIP) that greatly improves the quality of the generated images compared to previous versions of V1. It also includes an up-scaling model that improves the resolution of images by 4x.