app.py

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM
import gradio as gr

# Load DialoGPT model and tokenizer
tokenizer = AutoTokenizer.from_pretrained("microsoft/DialoGPT-large")
model = AutoModelForCausalLM.from_pretrained("microsoft/DialoGPT-large")

def respond(
    message,
    history: list[tuple[str, str]],
    system_message,
    max_tokens,
    temperature,
    top_p,
):
    # Prepare the context from history
    chat_history = ""
    for user_input, bot_response in history:
        if user_input:
            chat_history += f"User: {user_input}\n"
        if bot_response:
            chat_history += f"Bot: {bot_response}\n"

    # Append the new user message
    chat_history += f"User: {message}\n"

    # Tokenize the input
    input_ids = tokenizer.encode(chat_history, return_tensors="pt")

    # Generate response
    output_ids = model.generate(
        input_ids,
        max_length=max_tokens + len(input_ids[0]),
        temperature=temperature,
        top_p=top_p,
        pad_token_id=tokenizer.eos_token_id,
    )

    # Decode the output and get the response
    output = tokenizer.decode(output_ids[0], skip_special_tokens=True)

    # Extract the bot's response
    bot_response = output.split("User:")[-1].split("Bot:")[-1].strip()
    history.append((message, bot_response))  # Update history

    yield bot_response


"""
For information on how to customize the ChatInterface, peruse the gradio docs: https://www.gradio.app/docs/chatinterface
"""
demo = gr.ChatInterface(
    respond,
    additional_inputs=[
        gr.Textbox(value="You are a friendly Chatbot.", 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()