1

Hi,

NIM with llama-3-8b models are not getting up whereas other models like codelllama, llama-3.1 etc. we are able to use without any issues.

We have tried Getting Started - NVIDIA Docs this as well but not getting output.

When we removed network=host then it is asking for NGC API key, though nim cache path is downloaded and given in docker run.

(Attached screenshot , it is getting stuck here without any error even we are not able to kill by ctrl-c or ctrl-z)

image
image625×303 50.7 KB

Thanks in Advance.