I only have 1 account and I got it to tier 4 just this morning. Around an hour later, I received the o3-mini usage enabled notification.
In fact, I created a new API key on the tier 4 account, still no access.
same here. i did not get an email but i saw on tier plans that if i upgrade to tier 3 / 4 i will have access to o3 mini in the table. then i upraded and o3 mini access disappear?
so what do i do with these tokens i bought if i don’t have access to o3 mini
hi edwin. i have not received any emails for access but it stated on the api tiers page that there would be access @ tier 3 and 4… so i proceeded to purchase to tier 4. now no access. who can i reach out to for assistance
I’m experiencing an issue accessing the o3-mini model via API, even though I believe I meet all the prerequisites. Here are the details of my situation:
- My account is on usage tier 4.
- I received the API access email for
o1ando3-minithis morning from Nikunj Handa. - Despite this, when I try to use
o3-miniin my API requests, I get an error indicating that the model is not available.
Error code: 404 - {‘error’: {‘message’: ‘The model o3-mini does not exist or you do not have access to it.’, ‘type’: ‘invalid_request_error’, ‘param’: None, ‘code’: ‘model_not_found’}}
Is there a delay in access activation after receiving the email? Or is there any additional step I need to complete?
Thanks in advance!
You can see if the model has been granted to your organization at all.
- In your API account, on platform.openai.com, go to your personal icon at upper-right, and choose “profile”.
- Then on the selections in the left bar menu, choose “limits” under the category “organization”.
- Review your organization usage tier which is listed at the top there.
- Below the short list of AI models, expand “show more models”.
- Scroll on down to where any
o1entries appear such as “o1-mini”, and see if they are accompanied by the model you expect to access.
Then:
Although it might not be the root cause, one additional step that you can take yourself if you do see a rate limit provided for the desired model is to review “project limits” for a named project that you are using. Remember that in a project, you can set limits for models by name, either by “allow”, or “block”. This may take a re-save or refresh to populate a project with the latest models, or you can even create a new project.
The project selector is a bit odd - it is the upper-left next to your organization name.
The API keys of a project also might store a state that is not updated; creating a new one will be another step you can try.
“Default project” now even has these limits you can set and forget.
Ultimate test of utility: List the models with the models API endpoint and a new project and key. Ensure the new key is being used in your API call environment.
The link in the forum top bar “contact support” explains how to send a message for account configuration support. Remember that the language “rolling out” often means “not you yet”.
Me too, I have always been tier 5. I still cannot access the o1 model, and the newly released o3-mini model is also inaccessible. The limits page does not show these models either.
makes sense but the problem is. it showed in each tier o3 was available then once we upgraded it does not show in usage limits. this is the main issue.
I just spent $300 total paying for this. But I too have no access even though it says I am now tier 4. What is this B.S?
ok thanks for the update. I wish they would make these bugs more readily available so that people do not feel scammed after making huge purchases
that makes sense, hope this gets resolved soon
What the heck? It’s like they said to themselves, “if word got out that we would ever lift a finger and fix an account, then others might send us messages, too! Press send on the suggested bot answer that pops up. Best regards.”
Prompt engineer the AI-powered support people perhaps:
Account Issue: Account does not have o3-mini deployed, an error in configuration.
Background: Account is usage payment tier 4, qualifying for this model.
Justification: The bulk of tier 3 through tier 5 accounts are currently using this model.
Action Required: Provision the modelo3-minifor organization org-xxx, or escalate to a support tier that can perform the required account configuration.
Response: Only made when corrective action has been taken.
Here is the issue. I have narrowed it down for the Open Admins to see.
So I go to the rate limits guide
As per screenshot here. It shows
then there is a message View Limits
which is fair - but the problem is if you are in a lower limit eg. in my case Tier 4. It shows that there is no o3-mini access. So naturally one would follow the advice as per screenshot to upgrade to Tier 5 - but when upgraded there is no access to o3-mini. That’s the issue. We are advised that there is access to it so we naturally upgrade to only see there is no access to it.
I can see the overlook here by openai but a lot of us have paid big money to upgrade to only find no access to o3.
I understand in retrospect that the media release said ‘rollout’ to a selected few however this does not indicate this in the Tier Structure. Remember a lot of us are being advised from multiple sources on the net about this rollout and we naturally access the dashboard not the media release
A simple hello test using my paid ‘Personal’ openai account / key. This is what I get “o3-mini` does not exist or you do not have access to it.”
Am I supposed to subscribe to a different subscription?
Yeah, there seems to be some misalignment between the blog and the tier settings.
In the release they said o3-mini would be available for all tier 3-5, so I thought “Alright, let me add some more funds to upgrade to tier 3”.
After not getting access, i found in the documentation it says only tier 4-5.
Well, let’s hope it is just a bug and wait a bit more to see how it goes.
I am tier 4 and I get this error:
Solution: if you can’t access o3 mini, you can always ‘explore’ (meaning burn hard earned cash on) the 10x more expensive but less intelligent, pre-deepidyseek models…
Let them eat cake 
Same here : I can’t access access o3 … I’ll just continue using DeepSeek in LM Model for now. Seems to work fine … even better.
The “select developers on API usage tiers 3-5” that the blog post mentions were chosen on Weds, meaning, if you reached tier 3 or higher after that, you won’t have access today. It’s not automatic, and I’m sorry that this wasn’t clearer earlier.
And that said, being in one of these tiers does not guarantee access immediately. For safety and security reasons, I can’t disclose all the eligibility criteria—but the “select”, established developers within those tiers that receive access first are those that are known to be using the models in good ways.
We’ll be doing more passes next week and so forth—both in giving access to more developers within tiers 3-5 and adding any new developers who recently reached tier 3 (and you’ll receive an email once o3-mini’s been enabled for your account).