Hi there
I was wondering if there is a way to instruct the assistant with an image.
Lets say i have a predefined structure in this image and i want the assistant to be initialised knowing this image and next to the image a excerp of its contents. This way i hope it has a better chance to collect the data correctly from similar images which are posted in messages.
Is there a way to do that?
I cant use filesearch because its images which and i need json schema outputs.