Update app.py

app.py

@@ -11,41 +11,46 @@ with gr.Blocks() as demo:
     msg = gr.Textbox()
     clear = gr.Button("Clear")
 
-    def user(user_message, history: list):
-        """Appends the user message to the conversation history."""
-        return "", history + [{"role": "user", "content": user_message}]
+    def chat(user_message, history: list):
+        """
+        Consolidated function that:
+          1) Appends the user message to the conversation history
+          2) Sends the conversation history to the vLLM API
+          3) Streams back the assistant's response
+        """
 
-    def bot(history: list):
-        """Sends the conversation history to the vLLM API and streams the assistant's response."""
-        # Append an empty assistant message to history to fill in as we receive the response
+        # 1) Append the user message
+        history = history + [{"role": "user", "content": user_message}]
+
+        # 2) Prepare an empty assistant entry for streaming
         history.append({"role": "assistant", "content": ""})
 
+        # 3) Stream the assistant's response
         try:
-            # Create a chat completion with streaming enabled using the client
             completion = client.chat.completions.create(
-                model="llama-3.2-3B-instruct",  # Adjust the model name if needed
+                model="llama-3.2-3B-instruct",
                 messages=history,
                 stream=True
             )
-
-            # Iterate over the streamed response
             for chunk in completion:
-                # Access the delta content from the chunk
                 delta = chunk.choices[0].delta
                 content = getattr(delta, 'content', '')
                 if content:
-                    # Update the assistant's message with new content
                     history[-1]['content'] += content
-                    yield history
+                    # Yield both the cleared textbox ("") and updated history
+                    yield "", history
         except Exception as e:
             # Handle exceptions and display an error message
             history[-1]['content'] += f"\n[Error]: {str(e)}"
-            yield history
+            yield "", history
 
-    # Set up the Gradio interface components
-    msg.submit(user, [msg, chatbot], [msg, chatbot], queue=False).then(
-        bot, chatbot, chatbot
+    # Wire up the single chat function to the UI
+    msg.submit(
+        fn=chat,               # single consolidated function
+        inputs=[msg, chatbot], # pass user message and current chatbot history
+        outputs=[msg, chatbot] # clear the message box and update the chatbot
     )
+
     clear.click(lambda: None, None, chatbot, queue=False)
 
 if __name__ == "__main__":