Model Config Error on VLLM?
#3
by
rodMetal
- opened
Anyone getting this error?
"ValueError: No supported config format found in nvidia/Llama-3.1-Nemotron-70B-Instruct"
I was getting the same error. I went and used this one instead, works just fine:
https://huggingface.co./nvidia/Llama-3.1-Nemotron-70B-Instruct-HF
Not sure what the difference really is between these 2 repos, but the -HF one seems to have the params (safetensors).