app.py

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

# Load tokenizer and model
tokenizer = AutoTokenizer.from_pretrained("meta-llama/Meta-Llama-3-8B")
model = AutoModelForCausalLM.from_pretrained("meta-llama/Meta-Llama-3-8B")

# Generation function
def generate_text(prompt):
    inputs = tokenizer(prompt, return_tensors="pt")
    with torch.no_grad():
        outputs = model.generate(
            **inputs,
            max_new_tokens=100,
            do_sample=True,
            temperature=0.7,
            top_p=0.9
        )
    return tokenizer.decode(outputs[0], skip_special_tokens=True)

# Gradio interface
iface = gr.Interface(
    fn=generate_text,
    inputs=gr.Textbox(lines=5, placeholder="Enter your prompt..."),
    outputs="text",
    title="LLaMA 3 Text Generator",
    description="Generate text using Meta-LLaMA 3 8B model"
)

if __name__ == "__main__":
    iface.launch()