app.py

import gradio as gr
from huggingface_hub import InferenceClient

def respond(message, history, system_message, max_tokens, temperature, top_p, selected_model):
    client = InferenceClient(selected_model)
    
    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

models = {
    "deepseek-ai/DeepSeek-Coder-V2-Instruct": "DeepSeek-Coder-V2-Instruct",
    "CohereForAI/c4ai-command-r-plus": "Cohere Command-R Plus",
    "meta-llama/Meta-Llama-3.1-8B-Instruct": "Meta-Llama-3.1-8B-Instruct",
    "bartowski/DeepSeek-V2-Chat-0628-GGUF": "DeepSeek-V2-Chat-0628-GGUF",
    "google/gemma-7b": "Gemma-7b",
    "openai-community/gpt2": "gpt2"
}

demo = gr.ChatInterface(
    respond,
    additional_inputs=[
        gr.Textbox(value="You are a friendly Chatbot.", label="시스템 메시지"),
        gr.Slider(minimum=1, maximum=2048, value=512, step=1, label="최대 새 토큰 수"),
        gr.Slider(minimum=0.1, maximum=4.0, value=0.7, step=0.1, label="온도"),
        gr.Slider(minimum=0.1, maximum=1.0, value=0.95, step=0.05, label="Top-p (핵 샘플링)"),
        gr.Radio(list(models.keys()), value=list(models.keys())[0], label="언어 모델 선택", info="사용할 언어 모델을 선택하세요")
    ],
)

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