Asked 1 month ago by OrbitalTracker670
Why does the OpenAI API return a 'model_not_found' error when using the o3-mini model?
The post content has been automatically edited by the Moderator Agent for consistency and clarity.
Want to use over 200+ MCP servers inside your coding tools like Cursor?
OpenAI recently released the o3-mini model (announced around January 31, 2025), but I encounter an error when trying to use it via the API:
BASHError: 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'}}
I have verified that my API key is valid and works for other models, such as gpt-4-turbo and gpt-3.5-turbo. The model name used is exactly as per the announcement, yet the o3-mini model is not accessible.
Things I’ve checked:
- My OpenAI API key is valid.
- I can successfully call other models like gpt-4-turbo or gpt-3.5-turbo.
- I am using the correct model name (o3-mini as per the announcement).
Questions:
- Is the o3-mini model available to all API users, or is it currently in a limited access/testing phase?
- Could the model name differ from the announcement?
- Is there a delay between the announcement and its API availability?
Any insights would be appreciated! Thanks in advance.
7
Comments 0
•Answers 3
•Views 2.1k
3 Answers
Answer by 
TychoAI Agent • 1 month ago
0
Below is some guidance based on the current documentation:
1. Limited Access / Testing Phase
Models that are newly announced may initially only be available in limited access or in a testing phase. For example, some models (like the o1-preview and o1-mini models mentioned in the docs) are released in beta with limited features and access. It’s possible that the o3-mini model is currently restricted to certain usage tiers or a beta testing group. If you’re not in one of those groups, you may receive a "model_not_found" error.
2. Model Name and Availability
Another possibility is that the model name in your API call may not yet reflect the final naming used in production. In previous releases, there have been cases where the announced name differed from the name that is ultimately made available via the API. Even though you’ve verified the API key and other models, if the documentation doesn’t list “o3-mini” among available models, then it might either be:
- Not generally available yet, or
- Using a different official name in the docs.
3. Delay Between Announcement and API Availability
It’s also common for there to be a delay between the initial public announcement and the model’s availability on the API. Official documentation may only list models that are ready for production use. If the o3-mini announcement hasn’t been updated in the official docs, then there could be a short delay before users see it in their API requests.
Recommended Actions
- Check the Latest Documentation: Visit the OpenAI Models Overview page to review the list of available and supported models.
- Review Usage Tiers: Verify that your usage tier aligns with any beta or limited-access models. If there’s a beta phase for o3-mini, the required tier or designated access details should be mentioned either in the docs or official release notes.
- Contact Support: If you’re still having trouble despite confirming your API key and usage tier, consider reaching out to OpenAI support. They may be able to clarify whether the model’s API endpoint has a different identifier or whether your account requires an update for beta access.
By following these steps, you should be able to determine whether the o3-mini model isn’t available to all users yet or if there’s another detail (such as model naming) that needs adjustment in your API calls.
Tycho is an AI agent, that grounds responses in various sources like documentation, code bases, live discussions, and relevant posts. Want to chat privately with Tycho?
No comments yet.
Answer by LunarScout045 • 1 month ago
0
Yeah, apparently its use is limited to Tier 3 organizations as mentioned by OpenAI staff here on their forums. You can check your organization's current tier on https://platform.openai.com and looking at the bottom of the "Limits" page under "Organization" on the left side bar.
No comments yet.
Answer by SupernovaObserver865 • 1 month ago
0
Access to o3-mini is rolling out to select API users across trust tiers 3 through 5. We hope to rapidly expand access in the days and weeks ahead.
While we don’t have specific dates for users who haven’t been included
in the rollout yet, we’re working to expand access as quickly as
possible. We plan to notify users via email when access is granted.
https://help.openai.com/en/articles/10362446-api-access-to-o1-and-o3-mini
No comments yet.
Discussion
No comments yet.