app.py

from transformers import AutoModelForCausalLM, AutoTokenizer

# Load model and tokenizer
model_name = "meta-llama/Llama-3.1-8B-Instruct"
model = AutoModelForCausalLM.from_pretrained(
    model_name,
    torch_dtype="auto",
    device_map="auto"
)
tokenizer = AutoTokenizer.from_pretrained(model_name)

# Initialize persistent conversation with a system message
system_message = {"role": "system", "content": "You are a helpful and harmless assistant. You are Qwen developed by Alibaba. You should think step-by-step."}
messages = [system_message]

# Chat loop to maintain persistence
while True:
    user_input = input("User: ")  # Get user input
    if user_input.lower() in {"exit", "quit"}:
        print("Chat session ended.")
        break

    # Append user message to the conversation history
    messages.append({"role": "user", "content": user_input})

    # Format the messages for the model
    text = tokenizer.apply_chat_template(
        messages,
        tokenize=False,
        add_generation_prompt=True
    )
    model_inputs = tokenizer([text], return_tensors="pt").to(model.device)

    # Generate response
    generated_ids = model.generate(
        **model_inputs,
        max_new_tokens=512
    )
    generated_ids = [
        output_ids[len(input_ids):] for input_ids, output_ids in zip(model_inputs.input_ids, generated_ids)
    ]

    response = tokenizer.batch_decode(generated_ids, skip_special_tokens=True)[0]

    # Append assistant's response to the conversation history
    messages.append({"role": "assistant", "content": response})

    # Display the assistant's response
    print(f"Assistant: {response}")