Hi, I need some help. I’m using an assistant configured with “Function Calling,” and the goal is to ask questions about various institute programs and receive the relevant information in return.
My problem is that instead of making a single call to the get_data
function with all the programs (and their codes) and all the requested data, the assistant splits the request into multiple calls, which slows down the response process.
For example, I need a single call like this:
get_data({"programs": "IL, IME, II", "search": "description, duration, campus"})
But instead, the assistant makes multiple separate calls, such as: get_data({"programs": "IL", "search": "description, duration, campus"})
get_data({"programs": "IME", "search": "description, duration, campus"})
get_data({"programs": "II", "search": "description, duration, campus"})
Or it may do something like this:
get_data({"programs": "IL, IME, II", "search": "description"})
get_data({"programs": "IL, IME, II", "search": "duration"})
get_data({"programs": "IL, IME, II", "search": "campus"})
This repetition of calls is slowing down my system.
Is there a way to limit the number of calls the assistant makes to the same function? (I have other functions that can be called in parallel without any issue).