本地部署中文版 llama.cpp

模型：johnlui/chinese-alpaca-7b-and-13b-quantized

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

修改 llama.cpp examples/alpaca.sh

./main -m ./models/llama-7b-hf/ggml-model-q4_0.bin \
 --color  -f prompts/alpaca.txt -ins \
 -c 2048 --temp 0.2 -t 7 -n 2048 --repeat_penalty 1.3