Looking for direction on developing a custom GPT model.

[distant prawn]

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.

[sleek steppe]

You can directly make a GPT call API using actions, it will call the API whenever needed, you can parse the schema and api endpoint in actions

[distant prawn]

Hi Thunder, thank you for your response. I was under the impression that Actions were a ChatGPT-specific feature.

I will look into this.

How do Actions scale? Do they take up token count?

[gloomy relic]

If it is for your company, I would actually recommend the Assistants API