Update app.py

app.py

@@ -13,19 +13,23 @@ system_prompt = {
     "content": "You are a helpful assistant to answer user queries."
 }
 
-# Main chat function with memory from Gradio (OpenAI-style history)
+# Main chat function with memory from Gradio
 def get_text_response(user_message, history):
-    # Convert Gradio message history (OpenAI format) + new user message
-    messages = [system_prompt] + history + [{"role": "user", "content": user_message}]
+    # Convert Gradio message history to OpenAI format
+    formatted_history = [{"role": "user" if i % 2 == 0 else "assistant", "content": msg} 
+                         for i, msg in enumerate(sum(history, []))]
 
-    # Stream response
+    # Add the latest user message
+    messages = [system_prompt] + formatted_history + [{"role": "user", "content": user_message}]
+
+    # Stream the response
     response = ""
     completion = client.chat.completions.create(
         model="nvidia/llama-3.1-nemotron-70b-instruct",
         messages=messages,
         temperature=0.5,
         top_p=1,
-        max_tokens=1024,
+        max_tokens=100,
         stream=True
     )
 
@@ -36,12 +40,12 @@ def get_text_response(user_message, history):
 
     return response
 
-# Gradio Chat UI
+# Gradio Chat Interface
 demo = gr.ChatInterface(
     fn=get_text_response,
     title="🧠 Nemotron 70B Assistant",
     theme="soft",
-    chatbot=gr.Chatbot(height=400, type="messages"),  # <-- important: type="messages"
+    #chatbot=gr.Chatbot(height=400),  # default type is compatible
     textbox=gr.Textbox(placeholder="Ask me anything...", container=False),
     examples=["How are you doing?", "What are your interests?", "Which places do you like to visit?"]
 )