Streaming works perfectly!

mainV2.py

@@ -38,7 +38,7 @@ if __name__ == "__main__":
             editable="user",
             scale=1
         )
-        input_box = gr.Textbox(submit_btn=True, stop_btn=True, max_lines=5, label="Chat Message", scale=0)
+        input_box = gr.Textbox(max_lines=5, label="Chat Message", scale=0)
         
         chatbot.undo(handle_undo, chatbot, [chatbot, input_box])
         chatbot.retry(handle_retry, chatbot, chatbot)
@@ -52,10 +52,10 @@ if __name__ == "__main__":
         ).then(
             model_manager.run,  # Generate and stream response
             inputs=chatbot,
-            outputs=chatbot,
+            outputs=[chatbot, input_box],
+            queue=True,
             show_progress="full",
             trigger_mode="always_last"
         )
-        input_box.submit(lambda: "", None, [input_box])
 
     demo.launch(share=True)

src/manager.py

@@ -7,9 +7,7 @@ import sys
 from src.tool_loader import ToolLoader
 from src.utils.suppress_outputs import suppress_output
 import logging
-from gradio import ChatMessage
-
-from src.utils.streamlit_interface import get_user_message, output_assistant_response
+import gradio as gr
 
 logger = logging.getLogger(__name__)
 handler = logging.StreamHandler(sys.stdout)
@@ -120,7 +118,8 @@ class GeminiManager:
                 "content": f"Error generating response: {e}"
             })
             logger.error(f"Error generating response: {e}")
-            return messages
+            yield messages, gr.update(interactive=True)
+            return
         logger.debug(f"Response: {response}")
         print("Response: ", response)
 
@@ -137,7 +136,7 @@ class GeminiManager:
                 "role": "assistant",
                 "content": response.text
             })
-            yield messages
+            yield messages, gr.update(interactive=False,)
 
         # Attach the function call response to the messages
         if response.candidates[0].content and response.candidates[0].content.parts:
@@ -152,5 +151,6 @@ class GeminiManager:
             calls = self.handle_tool_calls(response)
             messages.append(calls)
             yield from self.run(messages)
+            return
         print("Final messages: ", messages)
-        return messages
+        yield messages, gr.update(interactive=True)