NVIDIA Nemotron API returns 403 Authorization Failed – Models endpoint works

AI & Data Science NVIDIA Nemotron
1

Hello NVIDIA Community,

I am currently integrating NVIDIA Nemotron 3.5 Lightning into my Python-based AI research agent project.

I am experiencing an authorisation issue with the NVIDIA API.

Here are my test results:

  1. GET /v1/models works successfully and returns 84 available models.

  2. POST /v1/chat/completions consistently returns HTTP 403 Forbidden.

  3. Error message: {“status”:403,“title”:“Forbidden”,“detail”:“Authorization failed”}

  4. Model: nvidia/nemotron-3.5-lightning-30b-a3b

  5. Endpoint: https://integrate.api.nvidia.com/v1

Interestingly, I can successfully generate responses using the NVIDIA Playground, but the same model fails when called through Python using the OpenAI SDK.

Could someone help me understand whether this is related to API key permissions, hosted inference access or organisation settings?

Thank you!