app.py

import gradio as gr
from huggingface_hub import InferenceClient

# Initialize the Hugging Face Inference Client
client = InferenceClient(model="meta-llama/Meta-Llama-3.1-405B-FP8")

# Define the response generation function
def respond(message, history, system_message, max_tokens, temperature, top_p):
    messages = [{"role": "system", "content": system_message}]

    # Add previous messages to the conversation
    for val in history:
        if val[0]:
            messages.append({"role": "user", "content": val[0]})
        if val[1]:
            messages.append({"role": "assistant", "content": val[1]})

    # Add the new user message
    messages.append({"role": "user", "content": message})

    response = ""

    # Generate the response using the model
    for message in client.chat_completion(
        messages=messages,
        max_tokens=max_tokens,
        stream=True,
        temperature=temperature,
        top_p=top_p,
    ):
        token = message.choices[0].delta.content
        response += token
        yield response

# Define the ChatGPT-like interface
with gr.Blocks(css=".gradio-container {max-width: 900px; margin: auto;}") as demo:
    gr.Markdown("<h1 style='text-align: center;'>ChatGPT-like Interface</h1>")
    
    chatbot = gr.Chatbot(height=500)
    with gr.Row():
        with gr.Column(scale=6):
            msg = gr.Textbox(
                show_label=False,
                placeholder="Type your message here...",
            )
        with gr.Column(scale=1, min_width=100):
            send_btn = gr.Button("Send")

    with gr.Accordion("Settings", open=False):
        system_message = gr.Textbox(value="You are a friendly Chatbot.", label="System message")
        max_tokens = gr.Slider(minimum=1, maximum=2048, value=512, step=1, label="Max new tokens")
        temperature = gr.Slider(minimum=0.1, maximum=4.0, value=0.7, step=0.1, label="Temperature")
        top_p = gr.Slider(minimum=0.1, maximum=1.0, value=0.95, step=0.05, label="Top-p (nucleus sampling)")