Closes #4

Added site tracking
Can pick the model the use
Better handle llm model based on debug or not

llm_be/chat_backend/consumers_graph.py

@@ -160,6 +160,7 @@ class ChatState(TypedDict):
     prompt_type: Union[PromptType, None]
     response_generator: Any # AsyncGenerator or dict
     error: Union[str, None]
+    model_name: str
 
 
 # --- LangGraph Nodes ---
@@ -206,7 +207,10 @@ async def generation_node(state: ChatState) -> ChatState:
 
     # Feature Flag: Internet Access
     if prompt_type == PromptType.SEARCH:
-         if getattr(settings, "ALLOW_INTERNET_ACCESS", False):
+         # Check modelName first - if FAST, we skip search regardless of settings
+         if state.get("model_name") == "FAST":
+             pass
+         elif getattr(settings, "ALLOW_INTERNET_ACCESS", False):
              try:
                  search = DuckDuckGoSearchRun()
                  search_results = search.run(state["message"])
@@ -274,6 +278,7 @@ class ChatConsumerGraph(AsyncWebsocketConsumer):
         print("Text Data: ", text_data)
         if text_data:
             data = json.loads(text_data)
+            model = data.get("modelName", "Turbo")
             message = data.get("message", None)
             conversation_id = data.get("conversation_id", None)
             email = data.get("email", None)
@@ -324,7 +329,8 @@ class ChatConsumerGraph(AsyncWebsocketConsumer):
                     "moderation_label": None,
                     "prompt_type": None,
                     "response_generator": None,
-                    "error": None
+                    "error": None,
+                    "model_name": model
                 }
                 print("Initial State: ", initial_state)