Run Code Llama locally with llama.cpp
git clone https://github.com/ggerganov/llama.cpp
cd llama.cpp/
LLAMA_METAL=1 make
cd models/
curl -OL https://huggingface.co/TheBloke/CodeLlama-7B-Instruct-GGUF/resolve/main/codellama-7b-instruct.Q5_K_M.gguf
cd ../
./main -m ./models/codellama-13b-instruct.Q5_K_M.gguf --color -ins -c 2048 --top_k 10000 --temp 0.2 --repeat_penalty 1.1 -ngl 4