A web service that generates text prompts for images that can be used with stable diffusion, a technique for creating realistic images with AI. The service uses CLIP, a model that learns from natural language and vision data. The service is optimized for clip ViT-L/14, a variant of CLIP that uses vision transformers.
