import gradio as gr
from transformers import AutoModelForCausalLM, AutoTokenizer

# Подключаем модель и токенизатор
model_name = "distilgpt2"
model = AutoModelForCausalLM.from_pretrained(model_name)
tokenizer = AutoTokenizer.from_pretrained(model_name)

def respond(message, history, 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})

    input_text = "\n".join([msg["content"] for msg in messages])

    inputs = tokenizer(input_text, return_tensors="pt", truncation=True, padding=True)

    outputs = model.generate(
        inputs["input_ids"],
        max_length=max_tokens,
        temperature=temperature,
        top_p=top_p,
        do_sample=True,
    )

    response = tokenizer.decode(outputs[0], skip_special_tokens=True)
    response += "\nСделано больницей EMS штата Alta!"
    return response

# Интерфейс Gradio
demo = gr.Interface(
    fn=respond,
    inputs=[
        gr.Textbox(value="Здравствуйте. Отвечай кратко...", label="System message"),
        gr.Slider(minimum=1, maximum=2048, value=512, label="Max Tokens"),
        gr.Slider(minimum=0.1, maximum=4.0, value=0.7, label="Temperature"),
        gr.Slider(minimum=0.1, maximum=1.0, value=0.95, label="Top-p"),
    ],
    outputs="text",
)

demo.launch()