When requesting model list via /v1/models, are the list items filtered by what the caller has access to, e.g. based on API Usage Tiers, paid vs free, etc.?
You cannot use the API without having paid, and it would be an anomaly to be fixed by support if you have credits and are stuck at “free” tier. There are no more trial credits that would be in effect.
This is easy to answer by just trying:
Free
babbage-002
dall-e-2
dall-e-3
davinci-002
gpt-3.5-turbo
gpt-3.5-turbo-0125
gpt-3.5-turbo-1106
gpt-3.5-turbo-16k
gpt-3.5-turbo-instruct
gpt-3.5-turbo-instruct-0914
gpt-4o-mini
gpt-4o-mini-2024-07-18
text-embedding-3-large
text-embedding-3-small
text-embedding-ada-002
tts-1
tts-1-1106
tts-1-hd
tts-1-hd-1106
whisper-1
Paid up, and project set to only allow two models in limits:
{‘data’: [{‘id’: ‘gpt-3.5-turbo-0613’, ‘created’: 1686587434, ‘object’: ‘model’, ‘owned_by’: ‘openai’}, {‘id’: ‘gpt-4o-mini’, ‘created’: 1721172741, ‘object’: ‘model’, ‘owned_by’: ‘system’}], ‘object’: ‘list’}
gpt-3.5-turbo-0613
gpt-4o-mini
Thank you for your answer! Perhaps my question will be clearer if I mention a specific scenario. Right now, the new o1- models are not returned from v1/models response I’m getting.
Will they show up in the response of this particular endpoint for organizations which qualify for the API Usage Tier 5?
Thank you!
Yes, the “o1-xx” model names are listed, both the model name with a date and an alias without a date that points to the current (and only) version.
The models automatically populated into an app of mine, after I changed my own filter rule which was supposed to get just “gpt-” for chat models (which is kind of good, because o1 would break a lot of naive apps).
And then having fired this old thing up for non-streaming, time to get some expertise on its UI shortcomings…
2 Likes