E pa nije ovo tragično sporo nakon /set parameter num_thread 3 :)
$ ollama run --verbose qwen2.5:0.5b
>>> /set parameter num_thread 3
Set parameter 'num_thread' to '3'
>>> hello, how are you today?
Hello! I'm Qwen, the AI language model created by Alibaba Cloud. How can I assist you today?
total duration: 3.504683535s
load duration: 2.030979805s
prompt eval count: 36 token(s)
prompt eval duration: 537.777283ms
prompt eval rate: 66.94 tokens/s
eval count: 24 token(s)
eval duration: 846.582586ms
eval rate: 28.35 tokens/s