The latest LMs are multimodal. They can translate text to text, text to images, and even images to images!
