app.py

import gradio as gr
from huggingface_hub import InferenceClient

client = InferenceClient("meta-llama/Llama-3.1-8B-Instruct")

def respond(message, history):
    
    messages = [{"role": "system", "content": "You are a shopping assistant. Help the user decide what items to buy based on their price constraints and interests. You should always give at least two options in response to questions."}]
    
    if history:
        messages.extend(history)
        
    messages.append({"role": "user", "content": message})

    response = ""

    for message in client.chat_completion(
        model="meta-llama/Llama-3.1-8B-Instruct",
        temperature=0.4,
        messages=messages,
        max_tokens=350,
        stream=True
    ):
        token = message.choices[0].delta.content

        if token is not None:
            response += token
            yield response

chatbot = gr.ChatInterface(respond)

chatbot.launch(debug=True)