Update app.py

app.py

@@ -1,25 +1,24 @@
 import gradio as gr
 from openai import OpenAI
 
-# NVIDIA-compatible OpenAI client
 client = OpenAI(
     base_url="https://integrate.api.nvidia.com/v1",
     api_key="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"
 )
 
-# System prompt
 system_prompt = {"role": "system", "content": "You are a helpful assistant to answer user queries."}
 
-# Main chat function with memory
-def get_text_response(user_message, history):
-    # history is a list of {"role": ..., "content": ...}
+def get_text_response(user_message, history=None):
     if history is None:
         history = []
 
-    # Add the user message to the history
-    messages = [system_prompt] + history + [{"role": "user", "content": user_message}]
+    # Prepare messages for OpenAI API in the correct format
+    messages = [system_prompt]
+    for user_msg, assistant_msg in history:
+        messages.append({"role": "user", "content": user_msg})
+        messages.append({"role": "assistant", "content": assistant_msg})
+    messages.append({"role": "user", "content": user_message})
 
-    # Stream response
     response = ""
     completion = client.chat.completions.create(
         model="nvidia/llama-3.1-nemotron-70b-instruct",
@@ -35,19 +34,13 @@ def get_text_response(user_message, history):
         if delta and delta.content:
             response += delta.content
 
-    # Update chat history
-    history.append({"role": "user", "content": user_message})
-    history.append({"role": "assistant", "content": response})
+    history.append((user_message, response))
+    return history, history
 
-    return response, history
-
-# Gradio Chat UI
 demo = gr.ChatInterface(
     fn=get_text_response,
     title="🧠 Nemotron 70B Assistant",
     theme="soft",
-    chatbot=gr.Chatbot(height=400, type="messages"),
-    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?"]
 )