diff --git a/vllm/v1/engine/async_llm.py b/vllm/v1/engine/async_llm.py index 720e800f02d56..ba2d5345d674a 100644 --- a/vllm/v1/engine/async_llm.py +++ b/vllm/v1/engine/async_llm.py @@ -63,7 +63,7 @@ def __init__( stream_mode=True) # EngineCore (starts the engine in background process). - self.engine_core = EngineCoreClient( + self.engine_core = EngineCoreClient.make_client( vllm_config=vllm_config, executor_class=executor_class, usage_context=usage_context,