llama.cpp

common : add -m to bash completion for --model [no ci]

#15591

Merged