Use my own fine-tune model in chat mode

I created my own fine-tune model. But I saw in the playground that it can only be used in complete mode, but I need chat mode. Is there any way I can use it in chat mode?

If yes: Please tell me how to do it.
If no: Will there be such a possibility in the future? Is there any way to get around this limitation? (Create a chat structure, but using complete mode)?

Welcome to the OpenAI community @vitaliyishkevychdev

As of writing this, fine-tuning is only supported for base models ada, babbage, curie, davinci.

These models are accessible via the completions endpoint only, which means that the fine tune models will be accessible only there.

Thus you cannot use fine tuned model on chat completions endpoint.

1 Like