Update app.py

app.py

@@ -145,25 +145,18 @@ def initialize_llm():
     try:
         print("Starting LLM initialization...")
         
-        api_token = os.environ.get('HF_TOKEN')
+        api_token = os.environ.get('HUGGINGFACE_API_TOKEN')
         if not api_token:
             raise ValueError("HUGGINGFACE_API_TOKEN not found in environment variables")
-        print("API token found")
-
-        # Initialize with custom LLM
-        llm = CustomHuggingFaceInference(token=api_token)
-        
-        # Test the LLM
-        print("Testing LLM with a simple prompt...")
-        test_response = llm("Hello, please reply with a short greeting.")
-        print(f"Test response received: {test_response}")
 
+        llm = HuggingFaceHub(
+            repo_id="mistralai/Mistral-7B-Instruct-v0.1",
+            huggingfacehub_api_token=api_token
+        )
         return llm
 
     except Exception as e:
         print(f"Error initializing LLM: {str(e)}")
-        import traceback
-        print(f"Traceback: {traceback.format_exc()}")
         raise
 
         
@@ -472,90 +465,48 @@ def get_chat_list():
 
 @app.route("/api/chat", methods=["POST"])
 def chat():
-    print("Entering chat endpoint")  # Log entry point
-    
-    if not llm_chain:
-        print("Error: LLM chain not initialized")
-        return jsonify({
-            "success": False,
-            "response": "System is still initializing. Please try again in a moment."
-        })
-
     try:
-        # Log incoming request
         data = request.json
-        print(f"Received request data: {data}")
-        
         user_input = data.get("message", "")
         session_id = data.get("sessionId", "default")
-        print(f"Processing message for session {session_id}: {user_input[:100]}...")
 
-        # Get or create session
-        print("Creating chat session...")
-        session = ChatSession(session_id)
+        print(f"Received message: {user_input}")
+
+        # Prepare the prompt
+        prompt = f"""Please provide a helpful response to: {user_input}
 
+        Guidelines:
+        1. If code is needed, use ```python for code blocks
+        2. Be clear and concise
+        3. If you don't know something, say so
+        """
+
+        # Get response from LLM
         try:
-            # Add user message
-            print("Adding user message...")
-            session.add_message("user", user_input)
-            update_chat_metadata(session_id, user_input)
-
-            # Get memory variables
-            print("Getting memory variables...")
-            memory_vars = session.get_memory_variables()
-            print(f"Memory variables loaded: {str(memory_vars)[:200]}...")
-
-            # Generate response
-            print("Generating LLM response...")
-            raw_response = llm_chain.run(
-                user_request=user_input,
-                chat_history=memory_vars.get("chat_history", ""),
-                important_info="\n".join(session.important_info)
-            )
-            print(f"Raw response received: {raw_response[:200]}...")
-
-            # Extract important information
-            print("Extracting important info...")
-            new_important_info = extract_important_info(raw_response)
-            print(f"Extracted important info: {new_important_info}")
-            
-            for info in new_important_info:
-                session.add_important_info(info)
-
-            # Format the response
-            print("Formatting response...")
-            formatted_response = format_response(raw_response)
-            
-            # Store the response
-            print("Storing assistant response...")
-            session.add_message("assistant", formatted_response)
-
-            print("Successfully completed chat processing")
+            response = llm(prompt)
+            print(f"Generated response: {response[:100]}...")
+
             return jsonify({
-                "response": formatted_response,
                 "success": True,
-                "important_info": session.important_info
+                "response": response
             })
 
-        except Exception as inner_e:
-            print(f"Inner processing error: {str(inner_e)}")
-            import traceback
-            print(f"Inner error traceback: {traceback.format_exc()}")
-            raise
+        except Exception as e:
+            print(f"Error generating response: {str(e)}")
+            return jsonify({
+                "success": False,
+                "response": "Error generating response. Please try again."
+            })
 
     except Exception as e:
-        import traceback
-        error_trace = traceback.format_exc()
         print(f"Error in chat endpoint: {str(e)}")
-        print(f"Full error traceback: {error_trace}")
-        
         return jsonify({
-            "response": f"An error occurred: {str(e)}",
             "success": False,
-            "error_trace": error_trace
+            "response": "An error occurred. Please try again."
         })
 
 
+
 @app.route("/api/new-chat", methods=["POST"])
 def new_chat():
     """Create a new chat session"""