import gradio as gr
from huggingface_hub import InferenceClient

client = InferenceClient("thviet79/model-QA-medical")

def respond(message, history, system_message, max_tokens, temperature, top_p):
    # Assuming it's a question-answering model
    response = client.question_answering(
        inputs={"question": message, "context": system_message}
    )
    return response['answer']  # Extract the answer from the response

demo = gr.ChatInterface(
    respond,
    additional_inputs=[
        gr.Textbox(value="Context for medical questions.", 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"),
    ],
)

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