llama-cpp-python>=0.2.0
# Optional GPU acceleration support, depends on user's CUDA version
# Users can choose to install based on their CUDA version
# Example: pip install llama-cpp-python --extra-index-url https://jllllll.github.io/llama-cpp-python-cuBLAS-wheels/AVX2/cu121 
# installation instructions: https://github.com/1038lab/ComfyUI-JoyCaption/blob/main/llama_cpp_install/llama_cpp_install.md
# CUDA version: https://github.com/1038lab/ComfyUI-JoyCaption/blob/main/llama_cpp_install.md#cuda-version-support

