app.py

import gradio as gr
from transformers import pipeline, set_seed
import os



api_key = os.environ.get("LLama2Secret")
# Your Hugging Face API key


# Initialize the pipeline with the LLaMA 2-7B chat model
model_name = "meta-llama/Llama-2-7b-chat"
# model_name = "meta-llama/Llama-2-7b-chat"
chat_model = pipeline("text2text-generation", model=model_name, use_auth_token=api_key)

def ask_llama2(question):
    set_seed(42)
    responses = chat_model(question, max_length=50)
    return responses[0]['generated_text']

iface = gr.Interface(fn=ask_llama2, inputs="text", outputs="text", title="Chat with LLaMA 2")

iface.launch()