Generate images from text prompts
RAG on documentations for your agent
Dense Grounded Understanding of Images and Videos