diff --git a/vllm/v1/executor/ray_utils.py b/vllm/v1/executor/ray_utils.py
index 21403e1c0..67c5a58f7 100644
--- a/vllm/v1/executor/ray_utils.py
+++ b/vllm/v1/executor/ray_utils.py
@@ -108,7 +108,7 @@ try:
 
             if isinstance(output, AsyncModelRunnerOutput):
                 output = output.get_output()
-            if not get_pp_group().is_last_rank:
+            if not self._is_last_rank():
                 # Case where there are no scheduled requests
                 # but may still be finished requests.
                 assert not output or not output.req_ids
@@ -128,6 +128,9 @@ try:
         def _is_intermediate_tensors(self, output) -> bool:
             return isinstance(output, IntermediateTensors)
 
+        def _is_last_rank(self) -> bool:
+            return get_pp_group().is_last_rank
+
     ray_import_err = None
 
 except ImportError as e: