Adding llama-3.1-nemotron-70b-instruct

[David-Sola]

Currently it is not possible to add the model from Nvidia.
It would be great for this possibility as the api costs are quite low.

[mamoodi]

It doesn't work if you enable Advanced Options and set it as the model?
Here is the litellm docs: https://docs.litellm.ai/docs/providers/nvidia_nim

Unless it's not supported by litellm

[neubig]

Yes @David-Sola , you could access the model via any hosting API that supports the model. For instance OpenRouter should work