Llama 3.1 nemotron 70b instruct API access not working correctly

AI & Data Science NVIDIA Nemotron
,
1

I started a project over a year ago using the Llama 3.1 nemotron 70b model as the nvidia build site suggested. This was the code I used to generate a response from the model:

client = OpenAI(

    base_url = "https://integrate.api.nvidia.com/v1",

    api_key = MY_API_KEY

)

completion = client.chat.completions.create(

    model="nvidia/llama-3.1-nemotron-70b-instruct",

    messages=PREV_MESSAGES,

    temperature=0.6,

    top_p=0.85,

    max_tokens=1024,

    stream=True

)

There was other code involved if needed, but I believe the problem is within this section of code. Iā€™ve had to renew my API key twice in the past and I hadnā€™t messed with my project in a while. When I came back it was not working due to the API key expiring so I renewed it the same way I did the other 2 times, made sure Iā€™m using the same library versions as before, and didnā€™t change anything else. I received the following error code and was not able to diagnose the problem:

File ā€œC:\Users\isaac\AppData\Local\Packages\PythonSoftwareFoundation.Python.3.13_qbz5n2kfra8p0\LocalCache\local-packages\Python313\site-packages\openai_base_client.pyā€, line 1034, in request
raise self._make_status_error_from_response(err.response) from None
openai.NotFoundError: Error code: 404 - {ā€˜statusā€™: 404, ā€˜titleā€™: ā€˜Not Foundā€™, ā€˜detailā€™: ā€œFunction ā€˜9b96341b-9791-4db9-a00d-4e43aa192a39ā€™: Not found for account ā€˜xhOhm8Fuo43Dj1yZv-6EspU97tp_SVoh5tcZbgEz6-cā€™ā€}

1 Like
2

Hi @isaac78 , we no longer host the Llama 3.1 nemotron 70b model. Iā€™d suggest you use either Llama 3.3 Nemotron Super 49b or Nemotron nano 9b v2.