app.py

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

# Load your environment variables
hf_api_token = os.getenv("HF_API_TOKEN")

# Ensure you have access to the model and are authenticated
model_name = "meta-llama/Meta-Llama-3-8B-Instruct"
tokenizer = AutoTokenizer.from_pretrained(model_name, token=hf_api_token)
model = AutoModelForCausalLM.from_pretrained(model_name, token=hf_api_token)

def chatbot(input_text):
    inputs = tokenizer.encode(input_text, return_tensors="pt")
    outputs = model.generate(inputs, max_length=500)
    response = tokenizer.decode(outputs[0], skip_special_tokens=True)
    return response

# Create Gradio interface
iface = gr.Interface(fn=chatbot, inputs="text", outputs="text", title="LLaMA 3 Chatbot")

# Use the Gradio queue to handle multiple requests
iface.queue().launch()