Generate high-quality images based on the user's recent prompt words. Allows the user to specify elements in the image to avoid using the "--none" parameter at the end of the prompt word (e.g. "tall trees, daytime, --no rain"). Technical support: Stable Diffusion XL.