places asyncio for stopasynciteration bug

app.py

@@ -5,6 +5,7 @@ from huggingface_hub import login
 import os
 from threading import Thread
 from openai import OpenAI
+import asyncio
 
 TOKEN = os.getenv('HF_AUTH_TOKEN')
 login(token=TOKEN,
@@ -171,10 +172,17 @@ def check_cuda():
 first_time = True
 llm_mode = ""
 
-def bot_comms(input_text: str,
-              history: list,
-              temperature: float,
-              max_new_tokens: int):
+# Async generator function
+async def async_generator():
+    for i in range(5):
+        # Simulate an asynchronous operation
+        await asyncio.sleep(1)
+        yield i
+
+async def bot_comms(input_text: str,
+                    history: list,
+                    temperature: float,
+                    max_new_tokens: int):
     """
     The connection between gradio and the LLM's
     """
@@ -248,6 +256,16 @@ def bot_comms(input_text: str,
             outputs.append(text)
             yield "".join(outputs)
 
+# Integration in your existing code
+async def main():
+    async for value in async_generator():
+        print(value)
+        # Add your existing logic here, e.g., call bot_comms with appropriate arguments
+        await bot_comms(input_text="example", history=[], temperature=0.5, max_new_tokens=128)
+
+# Rune async function
+asyncio.run(main())
+
 chatbot=gr.Chatbot(height=600, label="Loki AI")
 
 with gr.Blocks(fill_height=True) as demo: