Hello,
This might be a silly question.
I’m trying to understand, can I already use the GPT 4 turbo via API to create feature for a commercial platform?
I’m trying to understand is the “preview” version, is a real release version or only experiments version.
Thank you in advance!
Hi and welcome to the Dev Community!
Yes, the gpt-4-turbo model is available for use. The “preview” name was a bit more meaningful back in November/December when it released. Back then the model often had latency and reliability problems, but it’s improved significantly.
Essentially, you’re welcome to use it but OpenAI don’t quite feel comfortable yet explicitly saying its ready for production use.
Thanks for the answer!
That’s what I couldn’t really find, if it’s for production use by now.
Is a ‘production’ version is planned to be released anytime soon?
Another question in this relation, is if I’m using the preview modal for production use, can it be auto-updated to a newer version and possibly ‘break’ my feature?
Which is the latest stable modal to develop on?
Again thank you so much for your help!
I recommend taking a look at the models page in the API Documentation, the “gpt-4-turbo-preview” model will auto-update with each new release, whereas the others should remain unchanged.
GPT-4-Turbo is ready for production use in my opinion, just ensure you have good error handling in-place and you should be fine. You can also use the OpenAI Status Page/Alerts to get a heads-up when the API might be down.
Thank you kind sir 
very helpful