Update app.py

app.py

@@ -2,6 +2,7 @@ import gradio as gr
 import os
 import requests
 import threading
+from datetime import datetime
 from typing import List, Dict, Any
 
 # Get the Hugging Face API key from Spaces secrets
@@ -71,10 +72,19 @@ def query_model(model_name: str, messages: List[Dict[str, str]]) -> str:
     except Exception as e:
         return f"{model_name} error: {str(e)}"
 
-def respond(message: str, history: List[List[str]]) -> str:
-    """Handle sequential model responses with individual updates"""
+def respond(message: str, history: List[List[str]], session_id: str) -> str:
+    """Handle sequential model responses with session tracking"""
+    # Load session history
+    session = session_manager.load_session(session_id)
     messages = [{"role": "user", "content": message}]
     
+    # Store user message in session
+    session["history"].append({
+        "timestamp": datetime.now().isoformat(),
+        "type": "user",
+        "content": message
+    })
+    
     # Get first model's response
     response1 = query_model("Qwen2.5-Coder-32B-Instruct", messages)
     yield f"**Qwen2.5-Coder-32B-Instruct**:\n{response1}"
@@ -99,14 +109,32 @@ def respond(message: str, history: List[List[str]]) -> str:
     response3 = query_model("Llama3.3-70B-Instruct", messages)
     yield f"**Llama3.3-70B-Instruct**:\n{response3}"
 
-# Create the Gradio interface
-chat_interface = gr.ChatInterface(
-    respond,
-    title="Multi-LLM Collaboration Chat",
-    description="A group chat with Qwen2.5-72B, Llama3.3-70B, and Qwen2.5-Coder-32B",
-    examples=["How can I optimize Python code?", "Explain quantum computing basics"],
-    theme="soft"
-)
+# Create the Gradio interface with session management
+with gr.Blocks(title="Multi-LLM Collaboration Chat") as demo:
+    session_id = gr.State(session_manager.create_session)
+    
+    with gr.Row():
+        gr.Markdown("## Multi-LLM Collaboration Chat")
+        new_session_btn = gr.Button("🆕 New Session", variant="secondary")
+    
+    with gr.Row():
+        gr.Markdown("A group chat with Qwen2.5-72B, Llama3.3-70B, and Qwen2.5-Coder-32B")
+    
+    chat_interface = gr.ChatInterface(
+        respond,
+        examples=["How can I optimize Python code?", "Explain quantum computing basics"],
+        additional_inputs=[session_id]
+    )
+    
+    def create_new_session():
+        new_id = session_manager.create_session()
+        return new_id, None
+    
+    new_session_btn.click(
+        fn=create_new_session,
+        outputs=[session_id, chat_interface.chatbot],
+        show_progress=False
+    )
 
 if __name__ == "__main__":
     chat_interface.launch(share=True)