diff --git a/lmdeploy/serve/qos_engine/qos_engine.py b/lmdeploy/serve/qos_engine/qos_engine.py
index 1a311bc0f2..df1bf8e413 100644
--- a/lmdeploy/serve/qos_engine/qos_engine.py
+++ b/lmdeploy/serve/qos_engine/qos_engine.py
@@ -79,7 +79,7 @@ async def generate(self, request):
                 result_generator = self.engine.generate(
                     request.prompt[i],
                     request.session_id + i,
-                    True,  # always use stream to enable batching
+                    stream_response=True,  # always use stream for batching
                     sequence_start=True,
                     sequence_end=True,
                     request_output_len=request.max_tokens
@@ -119,7 +119,7 @@ async def generate(self, request):
             result_generator = self.engine.generate(
                 request.messages,
                 request.session_id,
-                True,  # always use stream to enable batching
+                stream_response=True,  # always use stream to enable batching
                 sequence_start=True,
                 sequence_end=True,
                 request_output_len=request.max_tokens