app.py

import gradio as gr
from huggingface_hub import InferenceClient

# Load the model from Hugging Face Hub
client = InferenceClient(model="tiiuae/falcon-7b-instruct")

# Chat completion function
def respond(message, history, system_message, max_tokens, temperature, top_p):
    messages = [{"role": "system", "content": system_message}]
    messages += history
    messages.append({"role": "user", "content": message})

    response = ""
    try:
        for chunk in client.chat_completion(
            messages=messages,
            max_tokens=max_tokens,
            stream=True,
            temperature=temperature,
            top_p=top_p,
        ):
            if hasattr(chunk.choices[0].delta, "content"):
                token = chunk.choices[0].delta.content
                response += token
                yield response
    except Exception as e:
        yield f"[Error] {e}"

# Gradio interface layout
with gr.Blocks() as demo:
    gr.Markdown("### 🧠 Falcon-7B-Instruct Chat UI — Powered by Hugging Face")
    
    with gr.Row():
        system_message = gr.Textbox(value="You are a helpful assistant.", label="System Prompt", lines=2)
    
    with gr.Row():
        message = gr.Textbox(placeholder="Ask something…", label="Your Message", lines=2)
    
    with gr.Row():
        max_tokens = gr.Slider(minimum=64, maximum=1024, value=256, step=64, label="Max Tokens")
        temperature = gr.Slider(minimum=0.1, maximum=1.0, value=0.7, step=0.1, label="Temperature")
        top_p = gr.Slider(minimum=0.1, maximum=1.0, value=0.9, step=0.1, label="Top-p (nucleus sampling)")

    chatbot = gr.Chatbot()
    state = gr.State([])

    submit = gr.Button("Send")

    def handle_submit(user_message, history, system_message, max_tokens, temperature, top_p):
        history = history + [[user_message, ""]]
        for updated_response in respond(user_message, history[:-1], system_message, max_tokens, temperature, top_p):
            history[-1][1] = updated_response
            yield history, history

    submit.click(
        handle_submit,
        inputs=[message, state, system_message, max_tokens, temperature, top_p],
        outputs=[chatbot, state],
    )

demo.launch()