changed for test reference

rag.py

@@ -77,7 +77,7 @@ def generate_stream(query:str,messages = [], model = "gpt-4o-mini", max_tokens =
         if stream:
             return llm_chain.stream({"context":context,"history":messages,"query":query})
         else:
-            return llm_chain.invoke({"context":context,"history":messages,"query":query})
+            return llm.invoke(query)
 
     except Exception as e:
         print(e)