Nemotron 51B too please
#1
by
nacs
- opened
Nvidia has a 51B that is a 'derivative' of this 70B.
Could you GGUF that too please: https://huggingface.co/nvidia/Llama-3_1-Nemotron-51B-Instruct
It's not supported sadly :( I wish
They're just teasing us with a long-awaited model between the 30b and 70b range, that can't be GGUF'd. π
Maybe convert to llama?
Waiting for something like this too. 48 GB of VRAM should be perfect for quantized 51B...