Hi NVIDIA team,
I’m getting a consistent 403 “Authorization failed” on all inference calls.
What works:
- GET https://integrate.api.nvidia.com/v1/models → HTTP 200 (key is valid)
What fails:
- POST https://integrate.api.nvidia.com/v1/chat/completions → HTTP 403
{“status”:403,“title”:“Forbidden”,“detail”:“Authorization failed”}
Models tested (all return the same 403):
- moonshotai/kimi-k3
- z-ai/glm-5.3
- z-ai/glm-5.3-flash
- deepseek-ai/deepseek-v4-flash-0731
This matches the known issue where personal organizations are missing the “Public API Endpoints” permission.
Could you please enable the “Public API Endpoints” permission for my personal organization?
Account details:
- Organization: Personal
- API Key prefix: nvapi-1DBmmADw…
Thank you.