llama.cpp

1cc0155d - server : tuning tests (#7388)

server : tuning tests (#7388) * server : don't pass temperature as string * server : increase timeout * tests : fix the fix 0.8f -> 0.8 ggml-ci * tests : set explicit temperature