Using multiple models in one application?

I have a small application using an older model right now. I was about to update to 3.5, but also got invited to try 4.

Is there a simple way to use multiple models?

The API requirements for text-davinci-003 (completion) and GPT-3.5/4 (chat completion) are not compatible. But, once you update to the new chat completion API, then it’s as easy as changing the model parameter in the request.

1 Like

As @wfhbrian said. I’ve got a few apps that test for a variable and connect to either the completions or chat endpoint, but that’s just stuff you need to do on your end in your code.

Is there something specific you’re trying to do?

Welcome to the forums!