Update app.py

app.py

@@ -7,6 +7,7 @@ import gradio as gr
 import torch
 from transformers import AutoModelForCausalLM, AutoTokenizer, TextIteratorStreamer
 from threading import Thread
+import asyncio
 
 TOKEN = os.environ["TOKEN"]
 
@@ -103,8 +104,6 @@ async def process_mention(message):
 
 generate("Hello, what is your name?")
 
-client.run(TOKEN)
-
 messages = []
 
 def refresh(new_message, markdown_text):
@@ -124,5 +123,10 @@ with gr.Blocks() as demo:
 
     submit_button.click(refresh, inputs=[message_input], outputs=[display_area])
 
-# Launch the app
-demo.launch()
+async def main():
+    loop = asyncio.get_event_loop()
+    gradio_task = loop.run_in_executor(None, demo.launch)
+
+    await client.run(TOKEN)
+
+asyncio.run(main())