feat(chat): enable streaming for chat responses

- [feat] Refactor handle_chat_submit to be a generator (chat_handler.py:handle_chat_submit())
- [feat] Implement iteration and yielding of partial responses (chat_handler.py:129-131)
- [refactor] Remove final response accumulation logic (chat_handler.py:124-127)
- [docs] Update docstring and comment for streaming (chat_handler.py:105,114)
- [feat] Add stream=True to chat_submit.click() (ui_components.py:75)
- [feat] Add stream=True to chat_input.submit() (ui_components.py:81)
- [docs] Update comment for chat event connection (ui_components.py:69)

chat_handler.py

@@ -102,15 +102,16 @@ def chat_respond(
 
 def handle_chat_submit(message, history, system_msg, model_name, max_tokens, temperature, top_p):
     """
-    Handle chat submission and manage conversation history.
+    Handle chat submission and manage conversation history with streaming.
     """
     if not message.strip():
-        return history, ""
+        yield history, ""
+        return
     
     # Add user message to history
     history = history + [{"role": "user", "content": message}]
     
-    # Generate response
+    # Generate response with streaming
     response_generator = chat_respond(
         message, 
         history[:-1],  # Don't include the current message in history for the function
@@ -121,12 +122,10 @@ def handle_chat_submit(message, history, system_msg, model_name, max_tokens, tem
         top_p
     )
     
-    # Get the final response
+    # Stream the assistant response token by token
     assistant_response = ""
     for partial_response in response_generator:
         assistant_response = partial_response
-    
-    # Add assistant response to history
-    history = history + [{"role": "assistant", "content": assistant_response}]
-    
-    return history, ""
+        # Update history with the current partial response and yield it
+        current_history = history + [{"role": "assistant", "content": assistant_response}]
+        yield current_history, ""

ui_components.py

@@ -66,19 +66,21 @@ def create_chat_tab(handle_chat_submit_fn):
         # Configuration tips below the chat
         create_chat_tips()
         
-        # Connect chat events
+        # Connect chat events with streaming enabled
         chat_submit.click(
             fn=handle_chat_submit_fn,
             inputs=[chat_input, chatbot_display, chat_system_message, chat_model_name, 
                    chat_max_tokens, chat_temperature, chat_top_p],
-            outputs=[chatbot_display, chat_input]
+            outputs=[chatbot_display, chat_input],
+            stream=True
         )
         
         chat_input.submit(
             fn=handle_chat_submit_fn,
             inputs=[chat_input, chatbot_display, chat_system_message, chat_model_name, 
                    chat_max_tokens, chat_temperature, chat_top_p],
-            outputs=[chatbot_display, chat_input]
+            outputs=[chatbot_display, chat_input],
+            stream=True
         )