Looking for advice on implementing middle-ware.

[flint crow]

Hello. I am looking to develop some functionality for my company's existing API that allows a GPT model to call a select few api calls. I imagined I would achieve this via middleware that watches for specific commands included in the GPT model's responses. I expect that I will need to fine-tune these into a custom model. Something like "/updatePreferences pref-name pref"

Is this a naïve approach? Is there a standardized way of doing this or an existing utility worth looking into? (Or even a better channel to post this in)

I welcome any reading materials people might suggest.

[cosmic mantle]

Hi! Firstly, the correct channel for this type of post is #1037561751362863144
Secondly, you can use function calling to achieve this. To set this up you can refer to the documentation of the assistants API, I recommend using the assistants API over the chat completions API since the assistants API is made for these kind of things.

Assistants API overview: https://platform.openai.com/docs/assistants/overview

Function calling: https://platform.openai.com/docs/assistants/tools/function-calling

[flint crow]

Hi Carrot, thanks! I was sent here from somewhere else, my apologies for posting in the wrong location.

Thank you for the resources. I appreciate it.

[cosmic mantle]

No problem

[elfin saffron]

sorry

already saw that was mentioned lol

but yeah function calling

[flint crow]

I was under the impression that was Specific to ChatGPT rather than the APIs. I'll have to look into it.