app.py

import gradio as gr
from huggingface_hub import InferenceClient

# Custom background CSS with styled title and panel
css = """
@import url('https://fonts.googleapis.com/css2?family=Noto+Sans+JP&family=Playfair+Display&display=swap');

body {
  background-image: url('https://cdn-uploads.huggingface.co/production/uploads/67351c643fe51cb1aa28f2e5/GdA9eNQKjOQjE6q47km3l.jpeg');
  background-size: cover;
  background-position: center;
  background-repeat: no-repeat;
  font-family: 'Noto Sans JP', sans-serif;
}
#chat-panel {
  background-color: rgba(255, 255, 255, 0.85);
  padding: 2rem;
  border-radius: 12px;
  max-width: 700px;
  height: 90vh;
  margin: auto;
  box-shadow: 0 0 12px rgba(0, 0, 0, 0.3);
  overflow-y: auto;
}
#chat-title {
  color: #2c3e50;
  font-family: 'Noto Sans', serif;
  font-size: 1.8rem;
  font-weight: bold;
  text-align: center;
  margin-bottom: 1rem;
}
"""

client = InferenceClient("HuggingFaceH4/zephyr-7b-beta")

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

with gr.Blocks(css=css) as demo:
    with gr.Column(elem_id="chat-panel"):
        gr.Markdown("## 🇯🇵 Japanese Tutor Chatbot", elem_id="chat-title")

        with gr.Accordion("⚙️ Settings", open=False):
            system_message = gr.Textbox(
                value="You are an expert Japanese tutor. Help users understand Japanese grammar, vocabulary, sentence structure, particles, and kanji readings. Reply clearly in English unless the user specifies otherwise.",
                label="System message"
            )
            max_tokens = gr.Slider(minimum=1, maximum=2048, value=512, step=1, label="Max new tokens")
            temperature = gr.Slider(minimum=0.1, maximum=4.0, value=0.7, step=0.1, label="Temperature")
            top_p = gr.Slider(minimum=0.1, maximum=1.0, value=0.95, step=0.05, label="Top-p (nucleus sampling)")

        gr.ChatInterface(
            respond,
            additional_inputs=[system_message, max_tokens, temperature, top_p]
        )

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