Comment by ngxson

Ahhh no sorry I forgot that the actual code controlling this is inside llama-model.cpp ; sorry for the misinfo, the -ngl only set to max by default if you're using Metal backend

(See the code in side llama_model_default_params())