Test

• try installing llama-cpp-python and llama-cpp-python[server] from pip ... WITH ggml-metal.metal (WITH Metal GPU support) file in python executable directory

Environment

• from previous test

Result

• llama-cpp-python[server] FAILS

Steps

##################################
# remove previous pip package

pip uninstall llama-cpp-python -y
pip list | grep llama
> [nothing found]



##################################
# fresh pip install - force ref

pip install llama-cpp-python --force-reinstall --upgrade --no-cache-dir
pip install 'llama-cpp-python[server]''


pip list | grep llama
>  llama-cpp-python  0.1.59


##################################
# TEST

python3 -m llama_cpp.server --model $MODEL