Finetune function calling response, is it possible?

Hi guys

So I am currently using function calling (with gpt 3.5 16k) I was wondering if I can fine tune the model, so the function calling response is more accurate?

Please let me know, thank you!


The 16k model cannot currently be tuned, so that’s a off the table, also I do not believe there is any provision for including functions in the fine tuning schema currently.