Correct type hint

vllm/engine/llm_engine.py

@@ -1394,7 +1394,7 @@ def _process_model_outputs(self, ctx: SchedulerContext) -> None:
         return None
 
     def _advance_to_next_step(
-            self, output: List[SamplerOutput],
+            self, output: SamplerOutput,
             seq_group_metadata_list: List[SequenceGroupMetadata],
             scheduled_seq_groups: List[ScheduledSequenceGroup]) -> None:
         """Given model output from a single run, append the tokens to the