| import subprocess | |
| # commented because the existing llama-cpp-python package was renoved fron requirements.txt | |
| subprocess.run("pip uninstall -y llama-cpp-python", shell=True) | |
| install_command = "CMAKE_ARGS='-DGGML_CUDA=on -DCUDA_PATH=/usr/local/cuda-12.2 -DCUDAToolkit_ROOT=/usr/local/cuda-12.2 -DCUDAToolkit_INCLUDE_DIR=/usr/local/cuda-12.2/include -DCUDAToolkit_LIBRARY_DIR=/usr/local/cuda-12.2/lib64' FORCE_CMAKE=1 pip install llama-cpp-python --no-cache-dir" | |
| subprocess.run(install_command, shell=True) | |
| subprocess.run("uvicorn app:app --host 0.0.0.0 --port 7860", shell=True) |