You must log in or register to comment.
Trusting this as a blind person would be a terrible idea, unless you had someone nearby who could verify the LLMs output (at which point they could just describe the image to you). At least it’s local.
I am not an AI advocate. But this will be cool if implemented well.