Do we need to download your Llama.cpp fork off Github?

#1
by AIGUYCONTENT - opened

As the title states...

Or can I just download the model and run it in Oobabooga?

Was in the model instructions: First, download and compile my Modified llama.cpp-b4139 v0.2. Compile it, then run
https://github.com/ymcki/llama.cpp-b4139
./llama-cli -m ~/Llama-3_1-Nemotron-51B-Instruct.Q3_K_S.gguf -p 'You are a Europea

Sign up or log in to comment