run llama-7b model

2025年3月13日 root

download llama.cpp

shell

git clone https://github.com/ggerganov/llama.cpp.git
cd llama.cpp
make

download model

shell

mkdir models/7B
cd models/7B
wget https://huggingface.co/TheBloke/LLaMa-7B-GGML/resolve/main/llama-7b.ggmlv3.q4_0.bin

run model interactively

shell

./main -m models/7B/llama-7b.ggmlv3.q4_0.bin --color -ins

run model onece

shell

./main -m models/7B/llama-7b.ggmlv3.q4_0.bin  --color -p your_question