app.py

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

# Get the Hugging Face token from the environment variable
hf_token = os.getenv("HUGGINGFACE_HUB_TOKEN")

# Model name
model_name = "meta-llama/Llama-3.2-3B-Instruct"

# Load the model and tokenizer with the token
tokenizer = AutoTokenizer.from_pretrained(model_name, use_auth_token=hf_token)
model = AutoModelForCausalLM.from_pretrained(model_name, use_auth_token=hf_token)

def predict(input_text):
    # Tokenize input and generate text
    inputs = tokenizer(input_text, return_tensors="pt")
    outputs = model.generate(**inputs)
    return tokenizer.decode(outputs[0], skip_special_tokens=True)

# Create a Gradio interface
interface = gr.Interface(
    fn=predict,
    inputs=gr.Textbox(label="Input Text"),
    outputs=gr.Textbox(label="Generated Output"),
    title="Meta-LLaMA-3.1-8B-Instruct",
    description="Generate text using the meta-llama/Llama-3.1-8B-Instruct model."
)

# Launch the interface
interface.launch()