diff --git a/python/sglang/srt/entrypoints/EngineBase.py b/python/sglang/srt/entrypoints/EngineBase.py
index 42ecb12aa..5d3162afd 100644
--- a/python/sglang/srt/entrypoints/EngineBase.py
+++ b/python/sglang/srt/entrypoints/EngineBase.py
@@ -29,6 +29,7 @@ class EngineBase(ABC):
         bootstrap_port: Optional[Union[List[int], int]] = None,
         bootstrap_room: Optional[Union[List[int], int]] = None,
         data_parallel_rank: Optional[int] = None,
+        rid: Optional[Union[List[str], str]] = None,
     ) -> Union[Dict, Iterator[Dict]]:
         """Generate outputs based on given inputs."""
         pass
diff --git a/python/sglang/srt/entrypoints/engine.py b/python/sglang/srt/entrypoints/engine.py
index ebab42a8f..57fde5116 100644
--- a/python/sglang/srt/entrypoints/engine.py
+++ b/python/sglang/srt/entrypoints/engine.py
@@ -184,6 +184,7 @@ class Engine(EngineBase):
         bootstrap_port: Optional[Union[List[int], int]] = None,
         bootstrap_room: Optional[Union[List[int], int]] = None,
         data_parallel_rank: Optional[int] = None,
+        rid: Optional[Union[List[str], str]] = None,
     ) -> Union[Dict, Iterator[Dict]]:
         """
         The arguments of this function is the same as `sglang/srt/managers/io_struct.py::GenerateReqInput`.
@@ -218,6 +219,7 @@ class Engine(EngineBase):
             bootstrap_port=bootstrap_port,
             bootstrap_room=bootstrap_room,
             data_parallel_rank=data_parallel_rank,
+            rid=rid,
         )
         generator = self.tokenizer_manager.generate_request(obj, None)
 
@@ -264,6 +266,7 @@ class Engine(EngineBase):
         bootstrap_port: Optional[Union[List[int], int]] = None,
         bootstrap_room: Optional[Union[List[int], int]] = None,
         data_parallel_rank: Optional[int] = None,
+        rid: Optional[Union[List[str], str]] = None,
     ) -> Union[Dict, AsyncIterator[Dict]]:
         """
         The arguments of this function is the same as `sglang/srt/managers/io_struct.py::GenerateReqInput`.
@@ -300,6 +303,7 @@ class Engine(EngineBase):
             bootstrap_port=bootstrap_port,
             bootstrap_room=bootstrap_room,
             data_parallel_rank=data_parallel_rank,
+            rid=rid,
         )
         generator = self.tokenizer_manager.generate_request(obj, None)
 
@@ -315,6 +319,7 @@ class Engine(EngineBase):
         audio_data: Optional[MultimodalDataInputFormat] = None,
         video_data: Optional[MultimodalDataInputFormat] = None,
         dimensions: Optional[int] = None,
+        rid: Optional[Union[List[str], str]] = None,
     ) -> Dict:
         """
         The arguments of this function is the same as `sglang/srt/managers/io_struct.py::EmbeddingReqInput`.
@@ -326,6 +331,7 @@ class Engine(EngineBase):
             audio_data=audio_data,
             video_data=video_data,
             dimensions=dimensions,
+            rid=rid,
         )
         generator = self.tokenizer_manager.generate_request(obj, None)
         ret = self.loop.run_until_complete(generator.__anext__())
@@ -338,6 +344,7 @@ class Engine(EngineBase):
         audio_data: Optional[MultimodalDataInputFormat] = None,
         video_data: Optional[MultimodalDataInputFormat] = None,
         dimensions: Optional[int] = None,
+        rid: Optional[Union[List[str], str]] = None,
     ) -> Dict:
         """
         Asynchronous version of encode method.
@@ -351,6 +358,7 @@ class Engine(EngineBase):
             audio_data=audio_data,
             video_data=video_data,
             dimensions=dimensions,
+            rid=rid,
         )
         generator = self.tokenizer_manager.generate_request(obj, None)
         return await generator.__anext__()