#python -c "from llama_cpp import GGML_USE_CUBLAS; print(GGML_USE_CUBLAS)" return False

2 messages · Page 1 of 1 (latest)

spare cosmos
#

I have done everything instructed in the docs but unfortunately I get "False" as the output from the command above.

karmic summit
#

try:CMAKE_ARGS="-DLLAMA_CUBLAS=on" pip install llama-cpp-python --force-reinstall --upgrade --no-cache-dir ;rather than:set CMAKE_ARGS=-DLLAMA_CUBLAS=on