NVIDIA Public API Endpoints authorization request

AI & Data Science NVIDIA NIM Access/Accounts
1

Hello NVIDIA Support Team,

I am trying to use the NVIDIA hosted NIM API for personal development and testing.

I have verified that my NVIDIA API key is recognized:

GET https://integrate.api.nvidia.com/v1/models
→ HTTP 200 OK

However, inference requests are rejected:

POST https://integrate.api.nvidia.com/v1/chat/completions
→ HTTP 403 Forbidden

Response:
{“status”:403,“title”:“Forbidden”,“detail”:“Authorization failed”}

The model I am testing is:

moonshotai/kimi-k3

My account is a Personal Organization, and I do not see an option to assign the NVIDIA Public API Endpoints service or the Public API Endpoints User role.

Could you please check whether the NVIDIA Public API Endpoints entitlement/permission is enabled for my Personal Organization and enable it if necessary?

I am using the API for personal development and testing.

I have not included my API key for security reasons. I can provide my account email, organization information, API-key prefix, and request details through a private support channel if required.

Thank you.

2

Account access issues are not handled here through the forums.
Please check the pinned posts:

for details.

Thank you!

3

Hello NVIDIA Support Team,

I am a personal developer using NVIDIA Build. My API keys are active, but requests to https://integrate.api.nvidia.com/v1/chat/completions return HTTP 400 Bad Request.

I tested both z-ai/glm-5.3 and openai/gpt-oss-20b, and both return the same error.

Could you please check whether Public API Endpoints permission is enabled for my personal organization and enable it if necessary?

This is for personal, non-commercial development and testing.

Thank you.