How to Use CMake with Pipenv to Install CUDA-enabled Packages

[michaelsheka]

I’m looking for guidance on how to use Pipenv to install llama-cpp-python with CUDA support. The installation command using pip is:

CMAKE_ARGS="-DGGML_CUDA=on" pip install llama-cpp-python

Could someone help me with how to implement this in Pipenv?

Thanks in advance!

[matteius]

@michaelsheka have you tried CMAKE_ARGS="-DGGML_CUDA=on" pipenv install llama-cpp-python ?