import gradio as gr
from transformers import AutoTokenizer, AutoModelForCausalLM, pipeline
from peft import PeftModel

# Load base + LoRA model
base_model = "TinyLlama/TinyLlama-1.1B-Chat-v1.0"
lora_model = "Futuresony/future_12_10_2024"

tokenizer = AutoTokenizer.from_pretrained(base_model)
base = AutoModelForCausalLM.from_pretrained(base_model)
model = PeftModel.from_pretrained(base, lora_model)
model.eval()

# Create generation pipeline
generator = pipeline("text-generation", model=model, tokenizer=tokenizer)

# Define the chat function
def respond(message, history, system_message, max_tokens, temperature, top_p):
    prompt = system_message + "\n"
    for user, bot in history:
        prompt += f"User: {user}\nAssistant: {bot}\n"
    prompt += f"User: {message}\nAssistant:"

    response = generator(
        prompt,
        max_new_tokens=max_tokens,
        temperature=temperature,
        top_p=top_p,
        do_sample=True,
        return_full_text=False,
    )[0]["generated_text"]

    yield response.strip()

# Set up Gradio UI
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()