Update app.py

app.py

@@ -55,18 +55,29 @@ def stream_gemini_response(user_message: str, messages: list) -> Iterator[list]:
         thinking_complete = False
 
         # Add initial thinking message
+        #messages.append(
+        #    ChatMessage(
+        #        role="assistant",
+        #        content="",
+        #        metadata={"title": "⚙️ Thinking: *The thoughts produced by the model are experimental"}
+        #    )
+        #)
+
+        response_buffer = parts[1].text
+        print(f"\n=== Starting Response ===\n{response_buffer}")
+        
         messages.append(
             ChatMessage(
                 role="assistant",
-                content="",
-                metadata={"title": "⚙️ Thinking: *The thoughts produced by the model are experimental"}
+                content=response_buffer
             )
         )
+        
+        thinking_complete = True
 
         for chunk in response:
             parts = chunk.candidates[0].content.parts
             current_chunk = parts[0].text
-            # thinking_complete = True
 
             if len(parts) == 2 and not thinking_complete:
                 # Complete thought and start response
@@ -105,7 +116,7 @@ def stream_gemini_response(user_message: str, messages: list) -> Iterator[list]:
             else:
                 # Stream thinking
                 thought_buffer += current_chunk
-                print(f"\n=== Thinking Chunk ===\n{current_chunk}" + f" = {len(parts)} : {thinking_complete} =")
+                print(f"\n=== Thinking Chunk ===\n{current_chunk}")
 
                 messages[-1] = ChatMessage(
                     role="assistant",