OpenAI API adds organization and project service-tier controls
Administrators can restrict allowed service tiers; disallowed requests return a 400 error.
OpenAI's October 21, 2026 changelog says projects can now select allowed service tiers in Platform settings, with organization-wide restrictions taking precedence. No exact effective time was provided.
The policy covers default, flex and priority; fast is treated as priority, while Scale Tier is excluded. Requests using auto or omitting service_tier can still fail with a 400 error if their resolved tier is disallowed.
This is web Platform administration for API use, not a ChatGPT app feature. The announcement specifies no new fee, Chinese interface support or language-capability changes. It does not announce higher allowances or lower prices. Not tested here.
https://developers.openai.com/api/docs/guides/error-codes
https://developers.openai.com/api/docs/supported-countries
The policy covers default, flex and priority; fast is treated as priority, while Scale Tier is excluded. Requests using auto or omitting service_tier can still fail with a 400 error if their resolved tier is disallowed.
This is web Platform administration for API use, not a ChatGPT app feature. The announcement specifies no new fee, Chinese interface support or language-capability changes. It does not announce higher allowances or lower prices. Not tested here.
https://developers.openai.com/api/docs/guides/error-codes
https://developers.openai.com/api/docs/supported-countries
Usage conditions
- Hong Kong
- Hong Kong is absent from the official OpenAI API supported-region list; no regional expansion announced.
- Macau
- Macau is absent from the official OpenAI API supported-region list; no regional expansion announced.
- Taiwan
- Taiwan is API-supported; administrative access and usable tiers remain subject to organization/project settings. Individual accounts not tested.
Official source: OpenAI API changelog — October 21, 2026: service tier controls