import gradio as gr
from huggingface_hub import InferenceClient

"""
For more information on `huggingface_hub` Inference API support, please check the docs: https://huggingface.co/docs/huggingface_hub/v0.22.2/en/guides/inference
"""
client = InferenceClient("meta-llama/Llama-3.2-3B-Instruct")


def respond(
    model_name,
    message,
    history: list[tuple[str, str]],
    system_message,
    max_tokens,
    temperature,
    top_p,
):
    messages = [{"role": "system", "content": system_message}]

    for val in history:
        if val[0]:
            messages.append({"role": "user", "content": val[0]})
        if val[1]:
            messages.append({"role": "assistant", "content": val[1]})

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

    response = ""

    # Model selection based on the button click
    if model_name == "Llama":
        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
    elif model_name == "Chatgpt":
        response = "ChatGPT functionality is not yet implemented."
    elif model_name == "Claude":
        response = "Claude functionality is not yet implemented."
    else:
        response = "Model not recognized."

    return response


# CSS for styling the interface
css = """
body {
    background-color: #06688E; /* Dark background */
    color: white; /* Text color for better visibility */
}
.gr-button {
    background-color: #42B3CE !important; /* White button color */
    color: black !important; /* Black text for contrast */
    border: none !important;
    padding: 8px 16px !important;
    border-radius: 5px !important;
}
.gr-button:hover {
    background-color: #e0e0e0 !important; /* Slightly lighter button on hover */
}
.gr-slider-container {
    color: white !important; /* Slider labels in white */
}
"""

# Define the Gradio interface with buttons and model selection
def gradio_interface(model_name, message, history, system_message, max_tokens, temperature, top_p):
    return respond(model_name, message, history, system_message, max_tokens, temperature, top_p)


demo = gr.Interface(
    fn=gradio_interface,
    inputs=[
        gr.Textbox(value="Hello!", label="User Message"),
        gr.Textbox(value="You are a virtual health assistant. Your primary goal is to assist with health-related queries.", label="System Message", visible=False),
        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)"
        ),
        gr.Button("Chatgpt"),
        gr.Button("Llama"),
        gr.Button("Claude"),
    ],
    outputs="text",
    css=css,  # Pass the custom CSS here
)

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