from ollama import chat, ChatResponse model = "deepseek-r1:70b" response: ChatResponse = chat( model=model, messages=[ { 'role': 'user', 'content': 'Hello, how are you?', } ], stream=True ) for chunk in response: print(chunk['message']['content'], end='', flush=True) print("\n")