import ollama

# 1. Define your variables
system_prompt = "You are a helpful assistant with a witty personality."
context_prompt = "The current topic of conversation is quantum computing."
user_input = "Can you explain it like I'm five?"

# 2. Combine the prompts
# We merge the system and context for the 'system' role
combined_system = f"{system_prompt}\n\nContext: {context_prompt}"

# 3. Stream the response
stream = ollama.chat(
    model='llama3.2', # or your preferred model
    messages=[
        {'role': 'system', 'content': combined_system},
        {'role': 'user', 'content': user_input},
    ],
    stream=True,
)

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