Update app.py

app.py

@@ -100,6 +100,10 @@ with gr.Blocks(theme=gr.themes.Default(primary_hue="blue", secondary_hue="neutra
         and streams the response back to the chatbot.
         """
 
+        #Disable Qwen3 thinking
+        #if "Qwen3".lower() in current_selected_model:
+        #    system_prompt = system_prompt+" /no_think"
+        
         # The 'history' variable from Gradio contains the entire conversation.
         # We prepend the system prompt to this history to form the final payload.
         messages = [{"role": "system", "content": system_prompt}] + history
@@ -112,7 +116,8 @@ with gr.Blocks(theme=gr.themes.Default(primary_hue="blue", secondary_hue="neutra
         response_stream = ollama.chat(
             model=current_selected_model, # Use the dynamically selected model
             messages=messages,
-            stream=True
+            stream=True,
+            think=False
         )
 
         # Iterate through the stream, updating the placeholder with each new chunk.