app.py

import os
import gradio as gr
from openai import OpenAI


client = OpenAI(
    base_url="https://ueecxhqta9umllae.us-east-1.aws.endpoints.huggingface.cloud/v1/",
    api_key=os.environ.get("hf_token"),
)


def respond(
    지자체,
    제목,
    질문,
    max_tokens,
    temperature,
    top_p,
):
    messages = [{"role": "municipality", "content": 지자체}]
    messages.append({"role": "title", "content": 제목})
    messages.append({"role": "question", "content": 질문})
    response = ""

    chat_completion = client.chat.completions.create(
        model="tgi",
        messages=messages,
        stream=True,
        max_tokens=max_tokens,
        temperature=temperature,
        top_p=top_p,
    )

    for message in chat_completion:
        token = message.choices[0].delta.content

        response += token
        yield response


demo = gr.Interface(
    respond,
    inputs=["textbox", "textbox", "textbox"],
    outputs=["textbox"],
    additional_inputs=[
        gr.Slider(minimum=1, maximum=2048, value=512, step=1, label="Max new tokens"),
        gr.Slider(minimum=0.1, maximum=1.0, value=0.4, step=0.05, label="Temperature"),
        gr.Slider(
            minimum=0.1,
            maximum=1.0,
            value=0.90,
            step=0.05,
            label="Top-p (nucleus sampling)",
        ),
    ],
)


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