b

app.py

@@ -15,8 +15,7 @@ def format_messages(history: List[Tuple[str, str]], system_message: str, user_me
     messages.append({"role": "user", "content": str(user_message)})  # Convert user_message to string
     return messages
 
-def respond(message: str, history: List[Tuple[str, str]]) -> str:
-    # Default values for parameters
+async def respond(message: str, history: List[Tuple[str, str]]) -> str:
     system_message = "You are a helpful AI assistant."
     max_tokens = 1000
     temperature = 0.7
@@ -26,7 +25,7 @@ def respond(message: str, history: List[Tuple[str, str]]) -> str:
     response = ""
     
     try:
-        for msg in client.chat_completion(
+        async for msg in client.chat_completion(
             messages,
             max_tokens=max_tokens,
             stream=True,
@@ -38,9 +37,14 @@ def respond(message: str, history: List[Tuple[str, str]]) -> str:
                 if token is not None:
                     response += token
                     yield response
+        
+        if not response:  # Handle empty response case
+            yield "No response generated."
+            
     except Exception as e:
-        return f"Error: {str(e)}"
+        yield f"Error: {str(e)}"
 
+# Update the ChatInterface to use async function
 demo = gr.ChatInterface(
     fn=respond,
     title="Demo GPT-BI instruct",