llama.cpp
llama.cpp copied to clipboard
[Suggestion] Add parameter for setting openblas threads
Openblas deafults to some maximum available threads, but would probably not be the most optimal. In Openblas there is a function to set the number of threads, why not use this?
void openblas_set_num_threads(int num_threads);
Current workaround is to set an openblas environment variable.