# Import required libraries
import gradio as gr
from transformers import pipeline
import torch
import threading
import time
import tensorflow as tf

# Check GPU availability
print(torch.cuda.is_available())
print(tf.test.gpu_device_name())

# Initialize the text generation pipeline with the specified model
pipe = pipeline("text-generation", model="chargoddard/Yi-34B-Llama", device=0)

# Rate limiting parameters
rate_limit = 5  # Number of requests per second
last_request_time = 0

def respond(
    message,
    history: list[tuple[str, str]],
    system_message,
    max_tokens,
    temperature,
    top_p,
):
    global last_request_time

    # Apply rate limiting
    elapsed_time = time.time() - last_request_time
    if elapsed_time < 1.0 / rate_limit:
        time.sleep(1.0 / rate_limit - elapsed_time)

    last_request_time = time.time()

    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 = ""

    # Generate the response using the pipeline
    result = pipe(
        messages[-1]["content"],
        max_length=max_tokens,
        num_return_sequences=1,
        temperature=temperature,
        top_p=top_p,
    )

    response = result[0]['generated_text']
    yield response

# Gradio interface setup
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)",
        ),
    ],
)

# Launch the Gradio interface
demo.launch()