app.py

import gradio as gr
from transformers import AutoModelForCausalLM, AutoTokenizer
import os

# Set Hugging Face token
# Replace this with your Hugging Face token or set it as an environment variable in Spaces settings.
hf_token = os.getenv("HF_TOKEN")

# Load the Llama model with authentication
model_name = "meta-llama/Llama-3.2-3B-Instruct"
tokenizer = AutoTokenizer.from_pretrained(model_name, use_auth_token=hf_token)
model = AutoModelForCausalLM.from_pretrained(model_name, use_auth_token=hf_token)

# Define chatbot function
def chat_with_llama(user_input):
    inputs = tokenizer(user_input, return_tensors="pt")
    outputs = model.generate(inputs.input_ids, max_length=100, do_sample=True, temperature=0.7)
    response = tokenizer.decode(outputs[0], skip_special_tokens=True)
    return response

# Gradio interface
interface = gr.Interface(
    fn=chat_with_llama,
    inputs=gr.Textbox(lines=2, placeholder="Ask me anything!"),
    outputs=gr.Textbox(),
    title="Llama 3.2 3B Chatbot",
    description="A simple chatbot powered by Llama 3.2 3B."
)

# Launch the app
interface.launch()