feat: Remove response function yield

app.py

@@ -2,6 +2,7 @@ import os
 import gradio as gr
 from openai import AsyncOpenAI  #Use the Async client
 from dotenv import load_dotenv
+import re
 
 load_dotenv()
 
@@ -16,16 +17,7 @@ client = AsyncOpenAI(
     timeout=120.0
 )
 
-async def respond(
-    message,
-    history,
-    system_message,
-    max_tokens,
-    temperature,
-    top_p,
-):
-    
-    def sanitize_messages(history):
+def sanitize_messages(history):
         sanitized = []
         for m in history:
             # Check if it's the new dictionary format
@@ -37,30 +29,36 @@ async def respond(
                 if m[1]: sanitized.append({"role": "assistant", "content": m[1]})
         return sanitized
 
+def strip_think(text: str) -> str:
+    return re.sub(r"<think>.*?</think>", "", text, flags=re.DOTALL).strip()
+
+async def respond(
+    message,
+    history,
+    system_message,
+    max_tokens,
+    temperature,
+    top_p,
+):
+    
     messages = [{"role": "system", "content": system_message}]
     messages.extend(sanitize_messages(history))
     messages.append({"role": "user", "content": message})
 
-    response_text = ""
-
     try:
-        # Use await with the async client
-        stream = await client.chat.completions.create(
+        completion = await client.chat.completions.create(
             model="qwen/qwen3-32b",
             messages=messages,
             max_tokens=max_tokens,
             temperature=temperature,
             top_p=top_p,
-            stream=True,
         )
 
-        async for chunk in stream:
-            if chunk.choices[0].delta.content:
-                response_text += chunk.choices[0].delta.content
-                yield response_text
+        text = completion.choices[0].message.content
+        return strip_think(text)
 
     except Exception as e:
-        yield f"Error: {str(e)}"
+        return f"Error: {str(e)}"
 
 
 chatbot = gr.ChatInterface(