The interface for using the "Type": "image" with the ullm API differs significantly compared to OpenAI. It is necessary to refer to the source code to understand how to use it properly. Could you please add documentation to clarify this?
BTW, the design of ullm is very convenient and neat. I appreciate that I can choose to either provide a file path or submit my own encoded file data. I like it.
The interface for using the "Type": "image" with the ullm API differs significantly compared to OpenAI. It is necessary to refer to the source code to understand how to use it properly. Could you please add documentation to clarify this?
Code comparison:
BTW, the design of ullm is very convenient and neat. I appreciate that I can choose to either provide a file path or submit my own encoded file data. I like it.