A vision encoder is a necessary component for allowing many leading LLMs to be able to work with images uploaded by users.

Click to Do allows for quick tasks with anything that appears on the screen, and here's how to use it on your Copilot+ PC.