import os
import gradio as gr
from huggingface_hub import InferenceClient
import requests

# Define the file path and URL
model_filename = "AstroSage-8B-Q8.0.gguf"
model_url = "https://huggingface.co/AstroMLab/AstroSage-8B-GGUF/resolve/main/AstroSage-8B-Q8.0.gguf"

# Check if the model file exists locally; if not, download it
if not os.path.exists(model_filename):
    print(f"{model_filename} not found. Downloading...")
    response = requests.get(model_url, stream=True)
    response.raise_for_status()  # Check for any download errors
    with open(model_filename, "wb") as f:
        for chunk in response.iter_content(chunk_size=8192):
            f.write(chunk)
    print(f"Downloaded {model_filename} successfully.")

# Initialize the InferenceClient with the local file path
client = InferenceClient(model_filename)


def respond(
    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 = ""

    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 Chat Interface
demo = gr.ChatInterface(
    respond,
    additional_inputs=[
        gr.Textbox(
            value="Assume the role of AstroSage, a helpful chatbot designed to answer user queries about astronomy, astrophysics, and cosmology.",
            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()