Hi,
I have a scenario where I download images and pdfs, then put them through a OCR step and funnel the text into a openAI Assistant via the OpenAi Module.
I want to get rid of the extra OCR step.
4o is able to work with pdfs and images, at least it is in the ChatGPT UI.
But if I want to call my assistant over the OpenAI Module, I am only able to attach images.
How can I attach pdfs?
Thanks!