local_demos/transformers_demo.py

from transformers import AutoModelForCausalLM, AutoTokenizer
import torch

def generate_response(model, tokenizer, prompt, max_length=50):
    inputs = tokenizer(prompt, return_tensors="pt")
    outputs = model.generate(inputs.input_ids, max_length=max_length, num_return_sequences=1)
    response = tokenizer.decode(outputs[0], skip_special_tokens=True)
    return response

def main():
    model_name = "meta-llama/Llama-3.2-3B-Instruct"
    system_prompt = "You are a helpful assistant."

    tokenizer = AutoTokenizer.from_pretrained(model_name)
    model = AutoModelForCausalLM.from_pretrained(model_name)

    user_prompt = input("Enter your prompt: ")
    full_prompt = f"{system_prompt}\n{user_prompt}"

    response = generate_response(model, tokenizer, full_prompt)
    print("Response:", response)

if __name__ == "__main__":
    main()