This method, called "SketchAgent," uses a multimodal language model to turn natural language prompts into sketches in a few seconds.

For example, it can doodle a cat either on its own or through collaboration, drawing w/a human or incorporating text-based input to sketch each… pic.twitter.com/3qUBbVQbAe

— MIT CSAIL (@MIT_CSAIL) 4 juin 2025

This method, called "SketchAgent," uses a multimodal language model to turn natural language prompts into sketches in a few seconds. For example, it can doodle a cat either on its own or through collaboration, drawing w/a human or incorporating text-based input to sketch each

→ View original post on X — @mit_csail