app.py

import gradio as gr
from huggingface_hub import InferenceClient

# Load the Hugging Face model
client = InferenceClient("HuggingFaceH4/zephyr-7b-beta")

# Chatbot response function
def respond(
    message,
    history: list[tuple[str, str]],
    system_message,
    max_tokens,
    temperature,
    top_p,
):
    # Initialize message list with system prompt
    messages = [{"role": "system", "content": system_message}]

    # Add previous chat history
    for user_msg, assistant_msg in history:
        if user_msg:
            messages.append({"role": "user", "content": user_msg})
        if assistant_msg:
            messages.append({"role": "assistant", "content": assistant_msg})

    # Append latest user message
    messages.append({"role": "user", "content": message})

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

# Gradio UI with Jesse's professional identity
demo = gr.ChatInterface(
    fn=respond,
    additional_inputs=[
        gr.Textbox(
            value=(
                "Your name is Jesse. You are a highly professional, results-driven AI assistant. "
                "You are a CompTIA Tech+ certified technician with a bachelor's degree in Information Technology. "
                "You specialize in networking, IT support, cybersecurity, and troubleshooting. "
                "Do not use humor or casual language. Focus on providing accurate, concise, and actionable technical assistance. "
                "Always refer to yourself as Jesse."
            ),
            label="System message"
        ),
        gr.Slider(minimum=1, maximum=2048, value=512, step=1, label="Max new tokens"),
        gr.Slider(minimum=0.1, maximum=4.0, value=0.6, step=0.1, label="Temperature"),  # Slightly lowered for focus
        gr.Slider(minimum=0.1, maximum=1.0, value=0.85, step=0.05, label="Top-p (nucleus sampling)"),
    ],
    title="Jesse - CompTIA Tech+ Certified Technician",
    description="Speak with Jesse, a CompTIA Tech+ certified technician with a bachelor's degree in IT. Ask anything about networks, IT support, security, and diagnostics.",
    theme="default"
)

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