Update app.py

app.py

@@ -191,20 +191,19 @@ def respond(message, history: list, system_message, max_tokens, temperature, top
     messages_list.append({"role": "user", "content": message})
     
     # Generate the assistant's answer by streaming responses.
-response_text = ""
-for partial_response in client.chat_completion(
-    messages_list,
-    max_tokens=max_tokens,
-    stream=True,
-    temperature=temperature,
-    top_p=top_p,
-):
-    if partial_response.choices and partial_response.choices[0].delta:
-        token = partial_response.choices[0].delta.content
-        if token:
-            response_text += token
-            yield response_text
-
+    response_text = ""
+    for partial_response in client.chat_completion(
+        messages_list,
+        max_tokens=1024,
+        stream=True,
+        temperature=temperature,
+        top_p=top_p,
+    ):
+        if partial_response.choices and partial_response.choices[0].delta:
+            token = partial_response.choices[0].delta.content
+            if token:
+                response_text += token
+                yield response_text
 # --- Gradio Chat Interface ---
 demo = gr.ChatInterface(
     fn=respond,