Unable to fine tune on any model besides base models "ada, babbage, curie, davinci"

API Deprecations
,
1

Hi,
We have been working on fine tuning for ~6 months, and previously fine tuned on text-davinci-003

Recently, we have only been able to successfully fine tune on the curie and davinci models. Whenever we try to fine tune on another model we get the following error:
“Invalid base model: (model must be one of ada, babbage, curie, davinci) or a fine-tuned model created by your organization: XXX (HTTP status code: 400)”

We are stuck on only being able to fine tune on “ada, babbage, curie, davinci” and would like to know if there is any other models I can currently fine tune on besides these 4? If not, what is the time line on access to new models?

We have had to halt progress on our projects because davinci is not adequate and our existing fine tune models are not an adequate base.

2

We are working on safely enabling fine-tuning for GPT-4 and GPT-3.5 Turbo and expect this feature to be available later this year. [source]

1 Like
4

So are the base models, “ada, babbage, curie and davinci” the only options until the new releases?