API update: Engines -> Models

Hi everyone! We’re deprecating the term ‘engine’ in favor of ‘model’. Moving forward, API requests will reference a ‘model’ instead of an ‘engine’.

Most people already use these terms interchangeably, and we consistently hear that ‘model’ is more intuitive. If you’ve used a fine-tuned model, then you’re already familiar with using ‘model’ instead of ‘engine’ when making an API request. Engine listing is also being replaced by Model listing, which will consolidate both base and fine-tuned models in a single place.

We will maintain backward compatibility, but recommend updating your implementation as soon as you can to prevent future confusion.

You can find full details in the API Reference, but a call to the completions endpoint would now be:


response = openai.Completion.create(
  prompt=”Say hello world three times.”


curl https://api.openai.com/v1/completions \
-H "Content-Type: application/json" \
-H "Authorization: Bearer YOUR_API_KEY" \
-d '{"prompt": "Say hello world three times","model":"text-davinci-002"}'

openai api completions.create -m text-davinci-002 -p "Say hello world three times."

(An update to the node library will be released later today!)


Thanks for the update, I have 2 feedback.

I think the documentation needs a bit more detail. Usually, we have descriptions for request fields but we do not have descriptions for the response model. Example: what are permissions?

What is the naming pattern for the model name, I see sometimes it is subject-baseEngine-version but sometimes baseEngine-subject-version?

1 Like