Update main.py

main.py

@@ -168,9 +168,9 @@ async def start():
         )
         qa = ConversationalRetrievalChain.from_llm(
             streaming_llm,
+            memory=memory,
             chain_type="stuff",
             retriever=retriever_to_cache(),
-            memory=memory,
             return_source_documents=True,
             verbose=False
         )