app.py

import gradio as gr
from huggingface_hub import InferenceClient

# Connect to your model on Hugging Face Hub
client = InferenceClient("JK-TK/bible2")

# Define the response function
def respond(
    message,
    history: list[tuple[str, str]],
    system_message,
    max_tokens,
    temperature,
    top_p,
):
    # Construct a flat prompt: system + history + current message
    prompt = system_message.strip() + "\n\n"
    for user, assistant in history:
        prompt += f"User: {user}\nAssistant: {assistant}\n"
    prompt += f"User: {message}\nAssistant:"

    # Generate response from model using text-generation
    response = ""
    for token in client.text_generation(
        prompt,
        max_new_tokens=max_tokens,
        temperature=temperature,
        top_p=top_p,
        stream=True,
    ):
        response += token
        yield response


# Create the chat interface
demo = gr.ChatInterface(
    respond,
    additional_inputs=[
        gr.Textbox(value="You are a biblical AI assistant.", label="System message"),
        gr.Slider(minimum=1, maximum=1024, value=300, step=1, label="Max new tokens"),
        gr.Slider(minimum=0.1, maximum=2.0, value=0.8, step=0.1, label="Temperature"),
        gr.Slider(minimum=0.1, maximum=1.0, value=0.95, step=0.05, label="Top-p"),
    ],
)

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