ai_code.py

import os
import warnings
import torch
from transformers import pipeline, GenerationConfig
from transformers.utils import logging


warnings.filterwarnings("ignore")
logging.set_verbosity_error()


pipe = pipeline(
    "text-generation",
    model="TinyLlama/TinyLlama-1.1B-Chat-v1.0",
    device_map="auto",
    torch_dtype=torch.float16   
)


gen_config = GenerationConfig(
    max_new_tokens=40,
    max_length=None,
    do_sample=False  
)

print("Chatbot ready! Type quit() to exit.\n")

conversation = ""

while True:
    user_input = input("You: ")

    if user_input.lower() == "quit()":
        break


    conversation += f"<|user|>\n{user_input}\n<|assistant|>\n"

    
    prompt = f"""<|system|>
You are a helpful, friendly chatbot. Keep responses short and relevant.

{conversation}
"""

    print("Generating...")

    output = pipe(
        prompt,
        generation_config=gen_config,
        return_full_text=False
    )

    reply = output[0]["generated_text"].strip()

    print("Bot:", reply)

   
    conversation += reply + "\n"

  
    if len(conversation) > 2000:
        conversation = conversation[-2000:]