I analyzed the problem in depth. I get faster responses when I use the terminal, something is wrong with Ollama Python.

It only uses efficiency cores at 1.6Ghz :/, with terminal 5GHZ all cores.

i have:
i9 – 13980HX
24 Core – 32 Thread

I get faster responses when I use the terminal, something is wrong with Python.

Task Manager

How can ı fix that ?

import ollama

ollama_response = ollama.chat(
    model='llama3',
    stream=True,
    messages=[
        {
            'role': 'system',
            'content': '''You are an assistant with solid philosophical knowledge.''',
        },
        {
            'role': 'user',
            'content': 'what is the meaning of life?',
        },
    ],
    options={
        'num_gpu': 0,
        'num_thread': 32,
    }
)

for chunk in ollama_response:
    print(chunk['message']['content'], end='', flush=True)