403 Authorization Failed on /v1/chat/completions while /v1/models works

Hi NVIDIA Support,

I’m trying to use NVIDIA’s hosted NIM API with an API key generated from NVIDIA Build.

My API key is valid and begins with nvapi-.

This endpoint works successfully:

GET https://integrate.api.nvidia.com/v1/models

It returns the model list, including:

nvidia/nemotron-3.5-lightning-30b-a3b

However, requests to:

POST https://integrate.api.nvidia.com/v1/chat/completions

return:

403 Forbidden - Authorization failed

I am using my Personal organization and the API key was generated through NVIDIA Build.

Could you please check whether Public API Endpoints / hosted NIM inference access is enabled for my Personal organization?

I am using this only for development and prototyping of a personal AI assistant.

Thank you.