app.py

import gradio as gr
from huggingface_hub import InferenceClient

# 🔹 Initialize Hugging Face Inference Client
client = InferenceClient(model="one1cat/FineTunes_LLM_CFR_49")

def respond(message, history, system_message, max_tokens, temperature, top_p):
    """
    Generates responses using the fine-tuned CFR 49 model hosted on Hugging Face.
    """

    prompt = f"{system_message}\n\nUser: {message}\n\nAssistant:"

    response = ""

    try:
        for token in client.text_generation(
            prompt,
            max_new_tokens=max_tokens,
            temperature=temperature,
            top_p=top_p,
            stream=True,  # Enables real-time streaming output
        ):
            response += token
            yield response  # Stream response token by token

    except Exception as e:
        yield f"Error: {str(e)}"  # Handle potential API errors


# 🔹 Gradio Chat Interface
demo = gr.ChatInterface(
    respond,
    additional_inputs=[
        gr.Textbox(value="You are a CFR 49 regulatory compliance assistant.", 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=0.7, step=0.1, label="Temperature"),
        gr.Slider(minimum=0.1, maximum=1.0, value=0.95, step=0.05, label="Top-p (nucleus sampling)"),
    ],
)

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