diff --git a/src/_bentoml_impl/client/http.py b/src/_bentoml_impl/client/http.py
index 10c2cdda363..e265b97f028 100644
--- a/src/_bentoml_impl/client/http.py
+++ b/src/_bentoml_impl/client/http.py
@@ -327,12 +327,14 @@ def _call(
         kwargs: dict[str, t.Any],
         *,
         headers: t.Mapping[str, str] | None = None,
-    ) -> t.Any: ...
+    ) -> t.Any:
+        ...
 
     @abstractmethod
     def _get_stream(
         self, endpoint: ClientEndpoint, args: t.Any, kwargs: t.Any
-    ) -> t.Any: ...
+    ) -> t.Any:
+        ...
 
 
 class SyncHTTPClient(HTTPClient[httpx.Client]):
diff --git a/src/_bentoml_impl/serde.py b/src/_bentoml_impl/serde.py
index aff7ac07ded..95054dd3ef6 100644
--- a/src/_bentoml_impl/serde.py
+++ b/src/_bentoml_impl/serde.py
@@ -28,16 +28,20 @@ class Serde(abc.ABC):
     media_type: str
 
     @abc.abstractmethod
-    def serialize_model(self, model: IODescriptor) -> bytes: ...
+    def serialize_model(self, model: IODescriptor) -> bytes:
+        ...
 
     @abc.abstractmethod
-    def deserialize_model(self, model_bytes: bytes, cls: type[T]) -> T: ...
+    def deserialize_model(self, model_bytes: bytes, cls: type[T]) -> T:
+        ...
 
     @abc.abstractmethod
-    def serialize(self, obj: t.Any) -> bytes: ...
+    def serialize(self, obj: t.Any) -> bytes:
+        ...
 
     @abc.abstractmethod
-    def deserialize(self, obj_bytes: bytes) -> t.Any: ...
+    def deserialize(self, obj_bytes: bytes) -> t.Any:
+        ...
 
     async def parse_request(self, request: Request, cls: type[T]) -> T:
         """Parse a input model from HTTP request"""
diff --git a/src/_bentoml_sdk/api.py b/src/_bentoml_sdk/api.py
index 5a5d4016237..c000b82a513 100644
--- a/src/_bentoml_sdk/api.py
+++ b/src/_bentoml_sdk/api.py
@@ -87,10 +87,12 @@ def __attrs_post_init__(self) -> None:
             self.output_spec.media_type = DEFAULT_STREAM_MEDIA_TYPE
 
     @t.overload
-    def __get__(self: T, instance: None, owner: type) -> T: ...
+    def __get__(self: T, instance: None, owner: type) -> T:
+        ...
 
     @t.overload
-    def __get__(self, instance: object, owner: type) -> t.Callable[P, R]: ...
+    def __get__(self, instance: object, owner: type) -> t.Callable[P, R]:
+        ...
 
     def __get__(self: T, instance: t.Any, owner: type) -> t.Callable[P, R] | T:
         from pydantic.fields import FieldInfo
@@ -216,7 +218,8 @@ def _flatten_model_schema(model: type[IODescriptor]) -> dict[str, t.Any]:
 
 
 @t.overload
-def api(func: t.Callable[t.Concatenate[t.Any, P], R]) -> APIMethod[P, R]: ...
+def api(func: t.Callable[t.Concatenate[t.Any, P], R]) -> APIMethod[P, R]:
+    ...
 
 
 @t.overload
@@ -230,7 +233,8 @@ def api(
     batch_dim: int | tuple[int, int] = ...,
     max_batch_size: int = ...,
     max_latency_ms: int = ...,
-) -> t.Callable[[t.Callable[t.Concatenate[t.Any, P], R]], APIMethod[P, R]]: ...
+) -> t.Callable[[t.Callable[t.Concatenate[t.Any, P], R]], APIMethod[P, R]]:
+    ...
 
 
 def api(
diff --git a/src/_bentoml_sdk/service/dependency.py b/src/_bentoml_sdk/service/dependency.py
index bbcd4603385..d8a3c74716e 100644
--- a/src/_bentoml_sdk/service/dependency.py
+++ b/src/_bentoml_sdk/service/dependency.py
@@ -54,10 +54,12 @@ def get(
         return _dependent_cache[key]
 
     @t.overload
-    def __get__(self, instance: None, owner: t.Any) -> Dependency[T]: ...
+    def __get__(self, instance: None, owner: t.Any) -> Dependency[T]:
+        ...
 
     @t.overload
-    def __get__(self, instance: t.Any, owner: t.Any) -> T: ...
+    def __get__(self, instance: t.Any, owner: t.Any) -> T:
+        ...
 
     def __get__(self, instance: t.Any, owner: t.Any) -> Dependency[T] | T:
         if instance is None:
diff --git a/src/_bentoml_sdk/service/factory.py b/src/_bentoml_sdk/service/factory.py
index 79a732ed181..b0239195869 100644
--- a/src/_bentoml_sdk/service/factory.py
+++ b/src/_bentoml_sdk/service/factory.py
@@ -46,7 +46,8 @@
     HookF_ctx = t.TypeVar("HookF_ctx", bound=ContextFunc)
 
     class _ServiceDecorator(t.Protocol):
-        def __call__(self, inner: type[T]) -> Service[T]: ...
+        def __call__(self, inner: type[T]) -> Service[T]:
+            ...
 
 
 def with_config(
@@ -324,11 +325,13 @@ def serve_http(
 
 
 @t.overload
-def service(inner: type[T], /) -> Service[T]: ...
+def service(inner: type[T], /) -> Service[T]:
+    ...
 
 
 @t.overload
-def service(inner: None = ..., /, **kwargs: Unpack[Config]) -> _ServiceDecorator: ...
+def service(inner: None = ..., /, **kwargs: Unpack[Config]) -> _ServiceDecorator:
+    ...
 
 
 def service(inner: type[T] | None = None, /, **kwargs: Unpack[Config]) -> t.Any:
diff --git a/src/bentoml/_internal/bento/build_config.py b/src/bentoml/_internal/bento/build_config.py
index 1397bad5cff..5e7494f0378 100644
--- a/src/bentoml/_internal/bento/build_config.py
+++ b/src/bentoml/_internal/bento/build_config.py
@@ -791,7 +791,8 @@ def __init__(
             python: PythonOptions | dict[str, t.Any] | None = ...,
             conda: CondaOptions | dict[str, t.Any] | None = ...,
             models: list[ModelSpec | str | dict[str, t.Any]] | None = ...,
-        ) -> None: ...
+        ) -> None:
+            ...
 
     def __attrs_post_init__(self) -> None:
         use_conda = not self.conda.is_empty()
diff --git a/src/bentoml/_internal/client/__init__.py b/src/bentoml/_internal/client/__init__.py
index dc64b2cabaa..2be587975b4 100644
--- a/src/bentoml/_internal/client/__init__.py
+++ b/src/bentoml/_internal/client/__init__.py
@@ -87,17 +87,18 @@ async def async_wait_until_server_ready(
 
     @t.overload
     @staticmethod
-    def from_url(
-        server_url: str, *, kind: None | t.Literal["auto"] = ...
-    ) -> Client: ...
+    def from_url(server_url: str, *, kind: None | t.Literal["auto"] = ...) -> Client:
+        ...
 
     @t.overload
     @staticmethod
-    def from_url(server_url: str, *, kind: t.Literal["http"] = ...) -> HTTPClient: ...
+    def from_url(server_url: str, *, kind: t.Literal["http"] = ...) -> HTTPClient:
+        ...
 
     @t.overload
     @staticmethod
-    def from_url(server_url: str, *, kind: t.Literal["grpc"] = ...) -> GrpcClient: ...
+    def from_url(server_url: str, *, kind: t.Literal["grpc"] = ...) -> GrpcClient:
+        ...
 
     @staticmethod
     def from_url(
@@ -204,19 +205,22 @@ async def wait_until_server_ready(
     @classmethod
     async def from_url(
         cls, server_url: str, *, kind: None | t.Literal["auto"] = ...
-    ) -> AsyncGrpcClient | AsyncHTTPClient: ...
+    ) -> AsyncGrpcClient | AsyncHTTPClient:
+        ...
 
     @t.overload
     @classmethod
     async def from_url(
         cls, server_url: str, *, kind: t.Literal["http"] = ...
-    ) -> AsyncHTTPClient: ...
+    ) -> AsyncHTTPClient:
+        ...
 
     @t.overload
     @classmethod
     async def from_url(
         cls, server_url: str, *, kind: t.Literal["grpc"] = ...
-    ) -> AsyncGrpcClient: ...
+    ) -> AsyncGrpcClient:
+        ...
 
     @classmethod
     async def from_url(
@@ -322,19 +326,22 @@ def wait_until_server_ready(
     @classmethod
     def from_url(
         cls, server_url: str, *, kind: None | t.Literal["auto"] = ...
-    ) -> SyncGrpcClient | SyncHTTPClient: ...
+    ) -> SyncGrpcClient | SyncHTTPClient:
+        ...
 
     @t.overload
     @classmethod
     def from_url(
         cls, server_url: str, *, kind: t.Literal["http"] = ...
-    ) -> SyncHTTPClient: ...
+    ) -> SyncHTTPClient:
+        ...
 
     @t.overload
     @classmethod
     def from_url(
         cls, server_url: str, *, kind: t.Literal["grpc"] = ...
-    ) -> SyncGrpcClient: ...
+    ) -> SyncGrpcClient:
+        ...
 
     @classmethod
     def from_url(
diff --git a/src/bentoml/_internal/cloud/schemas/modelschemas.py b/src/bentoml/_internal/cloud/schemas/modelschemas.py
index 57e9a06ff79..aa905dba39c 100644
--- a/src/bentoml/_internal/cloud/schemas/modelschemas.py
+++ b/src/bentoml/_internal/cloud/schemas/modelschemas.py
@@ -343,9 +343,9 @@ class DeploymentTargetConfig:
     deployment_strategy: t.Optional[DeploymentStrategy] = attr.field(
         default=None
     )  # Specific
-    bento_deployment_overrides: t.Optional[ApiServerBentoDeploymentOverrides] = (
-        attr.field(default=None)
-    )
+    bento_deployment_overrides: t.Optional[
+        ApiServerBentoDeploymentOverrides
+    ] = attr.field(default=None)
     bento_request_overrides: t.Optional[BentoRequestOverrides] = attr.field(
         default=None
     )  # Put into image builder
diff --git a/src/bentoml/_internal/context.py b/src/bentoml/_internal/context.py
index c0162db0043..b9cd9e66702 100644
--- a/src/bentoml/_internal/context.py
+++ b/src/bentoml/_internal/context.py
@@ -81,12 +81,12 @@ def mutablecopy(self) -> Metadata:
 
 class ServiceContext:
     def __init__(self) -> None:
-        self._request_var: contextvars.ContextVar[ServiceContext.RequestContext] = (
-            contextvars.ContextVar("request")
-        )
-        self._response_var: contextvars.ContextVar[ServiceContext.ResponseContext] = (
-            contextvars.ContextVar("response")
-        )
+        self._request_var: contextvars.ContextVar[
+            ServiceContext.RequestContext
+        ] = contextvars.ContextVar("request")
+        self._response_var: contextvars.ContextVar[
+            ServiceContext.ResponseContext
+        ] = contextvars.ContextVar("response")
         # A dictionary for storing global state shared by the process
         self.state: dict[str, t.Any] = {}
 
diff --git a/src/bentoml/_internal/external_typing/starlette.py b/src/bentoml/_internal/external_typing/starlette.py
index 3045939dbf7..7c05b6fbeb0 100644
--- a/src/bentoml/_internal/external_typing/starlette.py
+++ b/src/bentoml/_internal/external_typing/starlette.py
@@ -10,7 +10,8 @@
     from starlette.types import Send as ASGISend
 
     class AsgiMiddleware(t.Protocol):
-        def __call__(self, app: ASGIApp, **options: t.Any) -> ASGIApp: ...
+        def __call__(self, app: ASGIApp, **options: t.Any) -> ASGIApp:
+            ...
 
     __all__ = [
         "AsgiMiddleware",
diff --git a/src/bentoml/_internal/external_typing/tensorflow.py b/src/bentoml/_internal/external_typing/tensorflow.py
index 60147ca4e4e..15a942acdfe 100644
--- a/src/bentoml/_internal/external_typing/tensorflow.py
+++ b/src/bentoml/_internal/external_typing/tensorflow.py
@@ -38,12 +38,14 @@
     from tensorflow.python.eager.function import ConcreteFunction
 
     class GenericFunction(t.Protocol):
-        def __call__(self, *args: t.Any, **kwargs: t.Any): ...
+        def __call__(self, *args: t.Any, **kwargs: t.Any):
+            ...
 
     # NOTE: future proof when tensorflow decided to remove EagerTensor
     # and enable eager execution on Tensor by default. This class only serves
     # as type fallback.
-    class EagerTensor(Tensor): ...
+    class EagerTensor(Tensor):
+        ...
 
 
 class SignatureMap(t.Mapping[str, t.Any], Trackable):
@@ -51,17 +53,22 @@ class SignatureMap(t.Mapping[str, t.Any], Trackable):
 
     _signatures: t.Dict[str, t.Union[ConcreteFunction, RestoredFunction, Function]]
 
-    def __init__(self) -> None: ...
+    def __init__(self) -> None:
+        ...
 
     def __getitem__(
         self, key: str
-    ) -> t.Union[ConcreteFunction, RestoredFunction, Function]: ...
+    ) -> t.Union[ConcreteFunction, RestoredFunction, Function]:
+        ...
 
-    def __iter__(self) -> t.Iterator[str]: ...
+    def __iter__(self) -> t.Iterator[str]:
+        ...
 
-    def __len__(self) -> int: ...
+    def __len__(self) -> int:
+        ...
 
-    def __repr__(self) -> str: ...
+    def __repr__(self) -> str:
+        ...
 
 
 # This denotes all decorated function with `tf.function`
diff --git a/src/bentoml/_internal/frameworks/diffusers.py b/src/bentoml/_internal/frameworks/diffusers.py
index e2a2a25c42e..5da553eb3bd 100644
--- a/src/bentoml/_internal/frameworks/diffusers.py
+++ b/src/bentoml/_internal/frameworks/diffusers.py
@@ -653,9 +653,9 @@ def get_runnable(bento_model: bentoml.Model) -> t.Type[bentoml.Runnable]:
             type[diffusers.DiffusionPipeline], _str2cls(pipeline_class)
         )
 
-    scheduler_class: str | type[diffusers.SchedulerMixin] | None = (
-        bento_options.scheduler_class
-    )
+    scheduler_class: str | type[
+        diffusers.SchedulerMixin
+    ] | None = bento_options.scheduler_class
 
     if scheduler_class and isinstance(scheduler_class, str):
         scheduler_class = t.cast(
@@ -673,9 +673,9 @@ def get_runnable(bento_model: bentoml.Model) -> t.Type[bentoml.Runnable]:
     low_cpu_mem_usage: bool | None = bento_options.low_cpu_mem_usage
     variant: str | None = bento_options.variant
     _torch_dtype: str | torch.dtype | None = bento_options.torch_dtype
-    device_map: str | dict[str, int | str | torch.device] | None = (
-        bento_options.device_map
-    )
+    device_map: str | dict[
+        str, int | str | torch.device
+    ] | None = bento_options.device_map
     load_pretrained_extra_kwargs = bento_options.load_pretrained_extra_kwargs
 
     support_lora = True if issubclass(pipeline_class, LoraLoaderMixin) else False
diff --git a/src/bentoml/_internal/frameworks/transformers.py b/src/bentoml/_internal/frameworks/transformers.py
index a6504d51208..9185f4fed58 100644
--- a/src/bentoml/_internal/frameworks/transformers.py
+++ b/src/bentoml/_internal/frameworks/transformers.py
@@ -56,14 +56,17 @@ class ModelDefaultMapping(t.TypedDict):
 
     class PreTrainedProtocol(t.Protocol):
         @property
-        def framework(self) -> str: ...
+        def framework(self) -> str:
+            ...
 
-        def save_pretrained(self, save_directory: str, **kwargs: t.Any) -> None: ...
+        def save_pretrained(self, save_directory: str, **kwargs: t.Any) -> None:
+            ...
 
         @classmethod
         def from_pretrained(
             cls, pretrained_model_name_or_path: str, *args: t.Any, **kwargs: t.Any
-        ) -> PreTrainedProtocol: ...
+        ) -> PreTrainedProtocol:
+            ...
 
     P = t.ParamSpec("P")
 
@@ -359,13 +362,15 @@ def delete_pipeline(task: str) -> None:
 @t.overload
 def load_model(
     bento_model: str | Tag | Model, **kwargs: t.Any
-) -> transformers.Pipeline: ...
+) -> transformers.Pipeline:
+    ...
 
 
 @t.overload
 def load_model(
     bento_model: str | Tag | Model, *args: t.Any, **kwargs: t.Any
-) -> TransformersPreTrained: ...
+) -> TransformersPreTrained:
+    ...
 
 
 def load_model(bento_model: str | Tag | Model, *args: t.Any, **kwargs: t.Any) -> t.Any:
diff --git a/src/bentoml/_internal/io_descriptors/base.py b/src/bentoml/_internal/io_descriptors/base.py
index 10412a53846..fb79a2056ea 100644
--- a/src/bentoml/_internal/io_descriptors/base.py
+++ b/src/bentoml/_internal/io_descriptors/base.py
@@ -79,7 +79,8 @@ class IODescriptor(ABC, _OpenAPIMeta, t.Generic[IOType]):
 
     if t.TYPE_CHECKING:
         # Populated by subclasses. Makes pyright happy.
-        def __init__(self, **kwargs: t.Any) -> None: ...
+        def __init__(self, **kwargs: t.Any) -> None:
+            ...
 
     __slots__ = (
         "_mime_type",
@@ -139,7 +140,8 @@ def sample(self, value: IOType) -> None:
     if t.TYPE_CHECKING:
 
         @classmethod
-        def from_sample(cls, sample: t.Any, **attrs: t.Any) -> t.Self: ...
+        def from_sample(cls, sample: t.Any, **attrs: t.Any) -> t.Self:
+            ...
 
     @abstractmethod
     def _from_sample(self, sample: t.Any) -> IOType:
diff --git a/src/bentoml/_internal/io_descriptors/multipart.py b/src/bentoml/_internal/io_descriptors/multipart.py
index 568281c68b1..a403529f63e 100644
--- a/src/bentoml/_internal/io_descriptors/multipart.py
+++ b/src/bentoml/_internal/io_descriptors/multipart.py
@@ -327,12 +327,14 @@ async def from_proto(self, field: pb.Multipart) -> dict[str, t.Any]:
     @t.overload
     async def _to_proto_impl(
         self, obj: dict[str, t.Any], *, version: t.Literal["v1"]
-    ) -> pb.Multipart: ...
+    ) -> pb.Multipart:
+        ...
 
     @t.overload
     async def _to_proto_impl(
         self, obj: dict[str, t.Any], *, version: t.Literal["v1alpha1"]
-    ) -> pb_v1alpha1.Multipart: ...
+    ) -> pb_v1alpha1.Multipart:
+        ...
 
     async def _to_proto_impl(
         self, obj: dict[str, t.Any], *, version: str
diff --git a/src/bentoml/_internal/io_descriptors/numpy.py b/src/bentoml/_internal/io_descriptors/numpy.py
index b909dcf36bd..02b53b51424 100644
--- a/src/bentoml/_internal/io_descriptors/numpy.py
+++ b/src/bentoml/_internal/io_descriptors/numpy.py
@@ -65,13 +65,15 @@
 @t.overload
 def dtypepb_to_npdtype_map(
     version: t.Literal["v1"] = ...,
-) -> dict[pb.NDArray.DType.ValueType, ext.NpDTypeLike]: ...
+) -> dict[pb.NDArray.DType.ValueType, ext.NpDTypeLike]:
+    ...
 
 
 @t.overload
 def dtypepb_to_npdtype_map(
     version: t.Literal["v1alpha1"] = ...,
-) -> dict[pb_v1alpha1.NDArray.DType.ValueType, ext.NpDTypeLike]: ...
+) -> dict[pb_v1alpha1.NDArray.DType.ValueType, ext.NpDTypeLike]:
+    ...
 
 
 @lru_cache(maxsize=2)
@@ -95,13 +97,15 @@ def dtypepb_to_npdtype_map(
 @t.overload
 def dtypepb_to_fieldpb_map(
     version: t.Literal["v1"] = ...,
-) -> dict[pb.NDArray.DType.ValueType, str]: ...
+) -> dict[pb.NDArray.DType.ValueType, str]:
+    ...
 
 
 @t.overload
 def dtypepb_to_fieldpb_map(
     version: t.Literal["v1alpha1"] = ...,
-) -> dict[pb_v1alpha1.NDArray.DType.ValueType, str]: ...
+) -> dict[pb_v1alpha1.NDArray.DType.ValueType, str]:
+    ...
 
 
 @lru_cache(maxsize=2)
@@ -121,13 +125,15 @@ def fieldpb_to_npdtype_map() -> dict[str, ext.NpDTypeLike]:
 @t.overload
 def npdtype_to_dtypepb_map(
     version: t.Literal["v1"] = ...,
-) -> dict[ext.NpDTypeLike, pb.NDArray.DType.ValueType]: ...
+) -> dict[ext.NpDTypeLike, pb.NDArray.DType.ValueType]:
+    ...
 
 
 @t.overload
 def npdtype_to_dtypepb_map(
     version: t.Literal["v1alpha1"] = ...,
-) -> dict[ext.NpDTypeLike, pb_v1alpha1.NDArray.DType.ValueType]: ...
+) -> dict[ext.NpDTypeLike, pb_v1alpha1.NDArray.DType.ValueType]:
+    ...
 
 
 @lru_cache(maxsize=2)
@@ -587,12 +593,14 @@ async def from_proto(self, field: pb.NDArray | bytes) -> ext.NpNDArray:
     @t.overload
     async def _to_proto_impl(
         self, obj: ext.NpNDArray, *, version: t.Literal["v1"]
-    ) -> pb.NDArray: ...
+    ) -> pb.NDArray:
+        ...
 
     @t.overload
     async def _to_proto_impl(
         self, obj: ext.NpNDArray, *, version: t.Literal["v1alpha1"]
-    ) -> pb_v1alpha1.NDArray: ...
+    ) -> pb_v1alpha1.NDArray:
+        ...
 
     async def _to_proto_impl(
         self, obj: ext.NpNDArray, *, version: str
diff --git a/src/bentoml/_internal/io_descriptors/pandas.py b/src/bentoml/_internal/io_descriptors/pandas.py
index 7c0c99a32c3..92752bad9f1 100644
--- a/src/bentoml/_internal/io_descriptors/pandas.py
+++ b/src/bentoml/_internal/io_descriptors/pandas.py
@@ -682,12 +682,14 @@ def process_columns_contents(content: pb.Series) -> dict[str, t.Any]:
     @t.overload
     async def _to_proto_impl(
         self, obj: ext.PdDataFrame, *, version: t.Literal["v1"]
-    ) -> pb.DataFrame: ...
+    ) -> pb.DataFrame:
+        ...
 
     @t.overload
     async def _to_proto_impl(
         self, obj: ext.PdDataFrame, *, version: t.Literal["v1alpha1"]
-    ) -> pb_v1alpha1.DataFrame: ...
+    ) -> pb_v1alpha1.DataFrame:
+        ...
 
     async def _to_proto_impl(
         self, obj: ext.PdDataFrame, *, version: str
@@ -1141,12 +1143,14 @@ async def from_proto(self, field: pb.Series | bytes) -> ext.PdSeries:
     @t.overload
     async def _to_proto_impl(
         self, obj: ext.PdSeries, *, version: t.Literal["v1"]
-    ) -> pb.Series: ...
+    ) -> pb.Series:
+        ...
 
     @t.overload
     async def _to_proto_impl(
         self, obj: ext.PdSeries, *, version: t.Literal["v1alpha1"]
-    ) -> pb_v1alpha1.Series: ...
+    ) -> pb_v1alpha1.Series:
+        ...
 
     async def _to_proto_impl(
         self, obj: ext.PdSeries, *, version: str
diff --git a/src/bentoml/_internal/marshal/dispatcher.py b/src/bentoml/_internal/marshal/dispatcher.py
index 1fbda49aa9a..f38ba339d63 100644
--- a/src/bentoml/_internal/marshal/dispatcher.py
+++ b/src/bentoml/_internal/marshal/dispatcher.py
@@ -143,9 +143,9 @@ def __init__(
         self.tick_interval = 0.001
 
         self._controller = None
-        self._queue: collections.deque[Job[T_IN, T_OUT]] = (
-            collections.deque()
-        )  # TODO(bojiang): maxlen
+        self._queue: collections.deque[
+            Job[T_IN, T_OUT]
+        ] = collections.deque()  # TODO(bojiang): maxlen
         self.get_batch_size = get_batch_size
         self._sema = shared_sema if shared_sema else NonBlockSema(1)
 
diff --git a/src/bentoml/_internal/models/model.py b/src/bentoml/_internal/models/model.py
index 8e1d0d6cb3e..422121c3bd6 100644
--- a/src/bentoml/_internal/models/model.py
+++ b/src/bentoml/_internal/models/model.py
@@ -639,10 +639,12 @@ def to_dict(self) -> t.Dict[str, t.Any]:
         return bentoml_cattr.unstructure(self)
 
     @overload
-    def dump(self, stream: io.StringIO) -> io.BytesIO: ...
+    def dump(self, stream: io.StringIO) -> io.BytesIO:
+        ...
 
     @overload
-    def dump(self, stream: None = None) -> None: ...
+    def dump(self, stream: None = None) -> None:
+        ...
 
     def dump(self, stream: io.StringIO | None = None) -> io.BytesIO | None:
         return yaml.safe_dump(self.to_dict(), stream=stream, sort_keys=False)  # type: ignore (bad yaml types)
diff --git a/src/bentoml/_internal/monitoring/api.py b/src/bentoml/_internal/monitoring/api.py
index 8eaf34fdafb..c8067b8694b 100644
--- a/src/bentoml/_internal/monitoring/api.py
+++ b/src/bentoml/_internal/monitoring/api.py
@@ -26,7 +26,8 @@ def monitor(
     name: str | t.Any,
     monitor_class: DefaultMonitor = ...,
     monitor_options: dict[str, t.Any] | None = ...,
-) -> t.Generator[DefaultMonitor, None, None]: ...
+) -> t.Generator[DefaultMonitor, None, None]:
+    ...
 
 
 @t.overload
@@ -35,7 +36,8 @@ def monitor(
     name: str | t.Any,
     monitor_class: str = ...,
     monitor_options: dict[str, t.Any] | None = ...,
-) -> t.Generator[MonitorBase[t.Any], None, None]: ...
+) -> t.Generator[MonitorBase[t.Any], None, None]:
+    ...
 
 
 @t.overload
@@ -44,7 +46,8 @@ def monitor(
     name: str | t.Any,
     monitor_class: None = ...,
     monitor_options: dict[str, t.Any] | None = ...,
-) -> t.Generator[MonitorBase[t.Any], None, None]: ...
+) -> t.Generator[MonitorBase[t.Any], None, None]:
+    ...
 
 
 @contextlib.contextmanager
diff --git a/src/bentoml/_internal/monitoring/base.py b/src/bentoml/_internal/monitoring/base.py
index d6b1ae2aca4..38a01ebab20 100644
--- a/src/bentoml/_internal/monitoring/base.py
+++ b/src/bentoml/_internal/monitoring/base.py
@@ -5,12 +5,12 @@
 import logging
 import typing as t
 
-MON_COLUMN_VAR: contextvars.ContextVar["dict[str, dict[str, str]] | None"] = (
-    contextvars.ContextVar("MON_COLUMN_VAR", default=None)
-)
-MON_DATAS_VAR: contextvars.ContextVar["dict[str, collections.deque[t.Any]] | None"] = (
-    contextvars.ContextVar("MON_DATAS_VAR", default=None)
-)
+MON_COLUMN_VAR: contextvars.ContextVar[
+    "dict[str, dict[str, str]] | None"
+] = contextvars.ContextVar("MON_COLUMN_VAR", default=None)
+MON_DATAS_VAR: contextvars.ContextVar[
+    "dict[str, collections.deque[t.Any]] | None"
+] = contextvars.ContextVar("MON_DATAS_VAR", default=None)
 
 DT = t.TypeVar("DT")
 MT = t.TypeVar("MT", bound="MonitorBase[t.Any]")
diff --git a/src/bentoml/_internal/runner/container.py b/src/bentoml/_internal/runner/container.py
index d64ef0cacf1..fc1694a1279 100644
--- a/src/bentoml/_internal/runner/container.py
+++ b/src/bentoml/_internal/runner/container.py
@@ -54,11 +54,13 @@ def create_payload(
 
     @classmethod
     @abc.abstractmethod
-    def to_payload(cls, batch: BatchType, batch_dim: int) -> Payload: ...
+    def to_payload(cls, batch: BatchType, batch_dim: int) -> Payload:
+        ...
 
     @classmethod
     @abc.abstractmethod
-    def from_payload(cls, payload: Payload) -> BatchType: ...
+    def from_payload(cls, payload: Payload) -> BatchType:
+        ...
 
     @t.overload
     @classmethod
@@ -67,7 +69,8 @@ def to_triton_payload(
         inp: tritongrpcclient.InferInput,
         meta: tritongrpcclient.service_pb2.ModelMetadataResponse.TensorMetadata,
         _use_http_client: t.Literal[False] = ...,
-    ) -> tritongrpcclient.InferInput: ...
+    ) -> tritongrpcclient.InferInput:
+        ...
 
     @t.overload
     @classmethod
@@ -76,7 +79,8 @@ def to_triton_payload(
         inp: tritonhttpclient.InferInput,
         meta: dict[str, t.Any],
         _use_http_client: t.Literal[True] = ...,
-    ) -> tritongrpcclient.InferInput: ...
+    ) -> tritongrpcclient.InferInput:
+        ...
 
     @classmethod
     def to_triton_payload(
@@ -127,25 +131,29 @@ def to_triton_http_payload(
     @abc.abstractmethod
     def batches_to_batch(
         cls, batches: t.Sequence[BatchType], batch_dim: int
-    ) -> tuple[BatchType, list[int]]: ...
+    ) -> tuple[BatchType, list[int]]:
+        ...
 
     @classmethod
     @abc.abstractmethod
     def batch_to_batches(
         cls, batch: BatchType, indices: t.Sequence[int], batch_dim: int
-    ) -> list[BatchType]: ...
+    ) -> list[BatchType]:
+        ...
 
     @classmethod
     @abc.abstractmethod
     def batch_to_payloads(
         cls, batch: BatchType, indices: t.Sequence[int], batch_dim: int
-    ) -> list[Payload]: ...
+    ) -> list[Payload]:
+        ...
 
     @classmethod
     @abc.abstractmethod
     def from_batch_payloads(
         cls, payloads: t.Sequence[Payload], batch_dim: int
-    ) -> tuple[BatchType, list[int]]: ...
+    ) -> tuple[BatchType, list[int]]:
+        ...
 
     @classmethod
     def get_batch_size(cls, batch: BatchType, batch_dim: int) -> int:
@@ -694,16 +702,16 @@ def register_builtin_containers():
 class AutoContainer(DataContainer[t.Any, t.Any]):
     @classmethod
     def to_payload(cls, batch: t.Any, batch_dim: int) -> Payload:
-        container_cls: t.Type[DataContainer[t.Any, t.Any]] = (
-            DataContainerRegistry.find_by_batch_type(type(batch))
-        )
+        container_cls: t.Type[
+            DataContainer[t.Any, t.Any]
+        ] = DataContainerRegistry.find_by_batch_type(type(batch))
         return container_cls.to_payload(batch, batch_dim)
 
     @classmethod
     def get_batch_size(cls, batch: Any, batch_dim: int) -> int:
-        container_cls: t.Type[DataContainer[t.Any, t.Any]] = (
-            DataContainerRegistry.find_by_batch_type(type(batch))
-        )
+        container_cls: t.Type[
+            DataContainer[t.Any, t.Any]
+        ] = DataContainerRegistry.find_by_batch_type(type(batch))
         return container_cls.get_batch_size(batch, batch_dim)
 
     @classmethod
@@ -718,7 +726,8 @@ def to_triton_payload(
         inp: tritongrpcclient.InferInput,
         meta: tritongrpcclient.service_pb2.ModelMetadataResponse.TensorMetadata,
         _use_http_client: t.Literal[False] = ...,
-    ) -> tritongrpcclient.InferInput: ...
+    ) -> tritongrpcclient.InferInput:
+        ...
 
     @t.overload
     @classmethod
@@ -727,7 +736,8 @@ def to_triton_payload(
         inp: tritonhttpclient.InferInput,
         meta: dict[str, t.Any],
         _use_http_client: t.Literal[True] = ...,
-    ) -> tritongrpcclient.InferInput: ...
+    ) -> tritongrpcclient.InferInput:
+        ...
 
     @classmethod
     def to_triton_payload(
@@ -752,18 +762,18 @@ def to_triton_payload(
     def batches_to_batch(
         cls, batches: t.Sequence[BatchType], batch_dim: int = 0
     ) -> tuple[BatchType, list[int]]:
-        container_cls: t.Type[DataContainer[t.Any, t.Any]] = (
-            DataContainerRegistry.find_by_batch_type(type(batches[0]))
-        )
+        container_cls: t.Type[
+            DataContainer[t.Any, t.Any]
+        ] = DataContainerRegistry.find_by_batch_type(type(batches[0]))
         return container_cls.batches_to_batch(batches, batch_dim)
 
     @classmethod
     def batch_to_batches(
         cls, batch: BatchType, indices: t.Sequence[int], batch_dim: int = 0
     ) -> list[BatchType]:
-        container_cls: t.Type[DataContainer[t.Any, t.Any]] = (
-            DataContainerRegistry.find_by_batch_type(type(batch))
-        )
+        container_cls: t.Type[
+            DataContainer[t.Any, t.Any]
+        ] = DataContainerRegistry.find_by_batch_type(type(batch))
         return container_cls.batch_to_batches(batch, indices, batch_dim)
 
     @classmethod
@@ -773,9 +783,9 @@ def batch_to_payloads(
         indices: t.Sequence[int],
         batch_dim: int = 0,
     ) -> list[Payload]:
-        container_cls: t.Type[DataContainer[t.Any, t.Any]] = (
-            DataContainerRegistry.find_by_batch_type(type(batch))
-        )
+        container_cls: t.Type[
+            DataContainer[t.Any, t.Any]
+        ] = DataContainerRegistry.find_by_batch_type(type(batch))
         return container_cls.batch_to_payloads(batch, indices, batch_dim)
 
     @classmethod
diff --git a/src/bentoml/_internal/runner/runnable.py b/src/bentoml/_internal/runner/runnable.py
index 1ccb1918bc8..403f41cd617 100644
--- a/src/bentoml/_internal/runner/runnable.py
+++ b/src/bentoml/_internal/runner/runnable.py
@@ -83,7 +83,8 @@ def method(
         batch_dim: tuple[int, int] | int = 0,
         input_spec: AnyType | tuple[AnyType, ...] | None = None,
         output_spec: AnyType | None = None,
-    ) -> RunnableMethod[T, P, R]: ...
+    ) -> RunnableMethod[T, P, R]:
+        ...
 
     @overload
     @staticmethod
@@ -94,7 +95,8 @@ def method(
         batch_dim: tuple[int, int] | int = 0,
         input_spec: AnyType | tuple[AnyType, ...] | None = None,
         output_spec: AnyType | None = None,
-    ) -> t.Callable[[t.Callable[t.Concatenate[T, P], R]], RunnableMethod[T, P, R]]: ...
+    ) -> t.Callable[[t.Callable[t.Concatenate[T, P], R]], RunnableMethod[T, P, R]]:
+        ...
 
     @staticmethod
     def method(
diff --git a/src/bentoml/_internal/runner/runner.py b/src/bentoml/_internal/runner/runner.py
index a9e64e74c97..d392df8eac9 100644
--- a/src/bentoml/_internal/runner/runner.py
+++ b/src/bentoml/_internal/runner/runner.py
@@ -130,11 +130,11 @@ def __getattr__(
             ...
 
         @t.overload
-        def __getattr__(
-            self, item: t.LiteralString
-        ) -> RunnerMethod[t.Any, P, t.Any]: ...
+        def __getattr__(self, item: t.LiteralString) -> RunnerMethod[t.Any, P, t.Any]:
+            ...
 
-        def __getattr__(self, item: str) -> t.Any: ...
+        def __getattr__(self, item: str) -> t.Any:
+            ...
 
     runner_methods: list[RunnerMethod[t.Any, t.Any, t.Any]]
     scheduling_strategy: type[Strategy]
diff --git a/src/bentoml/_internal/runner/runner_handle/__init__.py b/src/bentoml/_internal/runner/runner_handle/__init__.py
index fe9df40e915..828ae336b6b 100644
--- a/src/bentoml/_internal/runner/runner_handle/__init__.py
+++ b/src/bentoml/_internal/runner/runner_handle/__init__.py
@@ -22,7 +22,8 @@
 
 class RunnerHandle(ABC):
     @abstractmethod
-    def __init__(self, runner: AbstractRunner) -> None: ...
+    def __init__(self, runner: AbstractRunner) -> None:
+        ...
 
     @abstractmethod
     async def is_ready(self, timeout: int) -> bool:
@@ -34,7 +35,8 @@ def run_method(
         __bentoml_method: RunnerMethod[t.Any, P, R],
         *args: P.args,
         **kwargs: P.kwargs,
-    ) -> R: ...
+    ) -> R:
+        ...
 
     @abstractmethod
     async def async_run_method(
@@ -42,7 +44,8 @@ async def async_run_method(
         __bentoml_method: RunnerMethod[t.Any, P, R],
         *args: P.args,
         **kwargs: P.kwargs,
-    ) -> R: ...
+    ) -> R:
+        ...
 
     @abstractmethod
     def async_stream_method(
@@ -50,7 +53,8 @@ def async_stream_method(
         __bentoml_method: RunnerMethod[t.Any, P, R],
         *args: P.args,
         **kwargs: P.kwargs,
-    ) -> t.AsyncGenerator[R, None]: ...
+    ) -> t.AsyncGenerator[R, None]:
+        ...
 
 
 class DummyRunnerHandle(RunnerHandle):
diff --git a/src/bentoml/_internal/runner/strategy.py b/src/bentoml/_internal/runner/strategy.py
index 7d44278af92..e253bafadec 100644
--- a/src/bentoml/_internal/runner/strategy.py
+++ b/src/bentoml/_internal/runner/strategy.py
@@ -22,7 +22,8 @@ def get_worker_count(
         runnable_class: type[Runnable],
         resource_request: dict[str, t.Any] | None,
         workers_per_resource: int | float,
-    ) -> int: ...
+    ) -> int:
+        ...
 
     @classmethod
     @abc.abstractmethod
diff --git a/src/bentoml/_internal/server/base_app.py b/src/bentoml/_internal/server/base_app.py
index ff254d0da5c..aaf19cc9f70 100644
--- a/src/bentoml/_internal/server/base_app.py
+++ b/src/bentoml/_internal/server/base_app.py
@@ -35,7 +35,8 @@ def __init__(
 
     @property
     @abc.abstractmethod
-    def name(self) -> str: ...
+    def name(self) -> str:
+        ...
 
     @property
     def on_startup(self) -> list[LifecycleHook]:
diff --git a/src/bentoml/_internal/service/service.py b/src/bentoml/_internal/service/service.py
index c3bc14b8602..c41dfed797b 100644
--- a/src/bentoml/_internal/service/service.py
+++ b/src/bentoml/_internal/service/service.py
@@ -98,9 +98,9 @@ class Service:
     mount_apps: t.List[t.Tuple[ext.ASGIApp, str, str]] = attr.field(
         init=False, factory=list
     )
-    middlewares: t.List[t.Tuple[t.Type[ext.AsgiMiddleware], t.Dict[str, t.Any]]] = (
-        attr.field(init=False, factory=list)
-    )
+    middlewares: t.List[
+        t.Tuple[t.Type[ext.AsgiMiddleware], t.Dict[str, t.Any]]
+    ] = attr.field(init=False, factory=list)
 
     # gRPC related
     mount_servicers: list[tuple[ServicerClass, AddServicerFn, list[str]]] = attr.field(
diff --git a/src/bentoml/_internal/utils/__init__.py b/src/bentoml/_internal/utils/__init__.py
index c127d285a2b..87ddbc133f4 100644
--- a/src/bentoml/_internal/utils/__init__.py
+++ b/src/bentoml/_internal/utils/__init__.py
@@ -115,11 +115,13 @@ def add_experimental_docstring(f: t.Callable[P, t.Any]) -> t.Callable[P, t.Any]:
 
 
 @overload
-def first_not_none(*args: T | None, default: T) -> T: ...
+def first_not_none(*args: T | None, default: T) -> T:
+    ...
 
 
 @overload
-def first_not_none(*args: T | None) -> T | None: ...
+def first_not_none(*args: T | None) -> T | None:
+    ...
 
 
 def first_not_none(*args: T | None, default: None | T = None) -> T | None:
diff --git a/src/bentoml/container.py b/src/bentoml/container.py
index 00769a5f80b..cffbe1cebc0 100644
--- a/src/bentoml/container.py
+++ b/src/bentoml/container.py
@@ -61,7 +61,8 @@ def build(
     ssh: str | ArgType = ...,
     target: str | ArgType = ...,
     **kwargs: t.Any,
-) -> t.Any: ...
+) -> t.Any:
+    ...
 
 
 @t.overload
@@ -87,7 +88,8 @@ def build(
     metadata_file: PathType | None = ...,
     opt: tuple[str, ...] | dict[str, str | tuple[str, ...]] | None = ...,
     **kwargs: t.Any,
-) -> t.Any: ...
+) -> t.Any:
+    ...
 
 
 @t.overload
@@ -130,7 +132,8 @@ def build(
     target: str | None = ...,
     ulimit: str | dict[str, tuple[int, int]] | ArgType = ...,
     **kwargs: t.Any,
-) -> t.Any: ...
+) -> t.Any:
+    ...
 
 
 @t.overload
@@ -174,7 +177,8 @@ def build(
     snapshotter: str | None = ...,
     storage_driver: str | None = ...,
     **kwargs: t.Any,
-) -> t.Any: ...
+) -> t.Any:
+    ...
 
 
 @t.overload
@@ -273,7 +277,8 @@ def build(
     variant: str | None = ...,
     volume: str | tuple[str, str, str] | None = ...,
     **kwargs: t.Any,
-) -> t.Any: ...
+) -> t.Any:
+    ...
 
 
 @t.overload
@@ -355,7 +360,8 @@ def build(
     variant: str | None = ...,
     volume: str | tuple[str, str, str] | None = ...,
     **kwargs: t.Any,
-) -> t.Any: ...
+) -> t.Any:
+    ...
 
 
 @inject
diff --git a/src/bentoml/deployment.py b/src/bentoml/deployment.py
index 4906cd7ed29..ddbaf6a39af 100644
--- a/src/bentoml/deployment.py
+++ b/src/bentoml/deployment.py
@@ -39,7 +39,8 @@ def create(
     strategy: str | None = ...,
     envs: t.List[EnvItemSchema] | t.List[dict[str, t.Any]] | None = ...,
     extras: dict[str, t.Any] | None = ...,
-) -> DeploymentInfo: ...
+) -> DeploymentInfo:
+    ...
 
 
 @t.overload
@@ -51,7 +52,8 @@ def create(
     *,
     bento: Tag | str | None = ...,
     config_file: str | None = ...,
-) -> DeploymentInfo: ...
+) -> DeploymentInfo:
+    ...
 
 
 @t.overload
@@ -63,7 +65,8 @@ def create(
     *,
     bento: Tag | str | None = ...,
     config_dict: dict[str, t.Any] | None = ...,
-) -> DeploymentInfo: ...
+) -> DeploymentInfo:
+    ...
 
 
 @inject
@@ -136,7 +139,8 @@ def update(
     strategy: str | None = ...,
     envs: t.List[EnvItemSchema] | t.List[dict[str, t.Any]] | None = ...,
     extras: dict[str, t.Any] | None = ...,
-) -> DeploymentInfo: ...
+) -> DeploymentInfo:
+    ...
 
 
 @t.overload
@@ -149,7 +153,8 @@ def update(
     *,
     bento: Tag | str | None = ...,
     config_file: str | None = ...,
-) -> DeploymentInfo: ...
+) -> DeploymentInfo:
+    ...
 
 
 @t.overload
@@ -162,7 +167,8 @@ def update(
     *,
     bento: Tag | str | None = ...,
     config_dict: dict[str, t.Any] | None = ...,
-) -> DeploymentInfo: ...
+) -> DeploymentInfo:
+    ...
 
 
 @inject
@@ -235,7 +241,8 @@ def apply(
     *,
     bento: t.Optional[t.Union[Tag, str]] = ...,
     config_dict: t.Optional[dict[str, t.Any]] = ...,
-) -> DeploymentInfo: ...
+) -> DeploymentInfo:
+    ...
 
 
 @t.overload
@@ -248,7 +255,8 @@ def apply(
     *,
     bento: t.Optional[t.Union[Tag, str]] = ...,
     config_file: t.Optional[str] = ...,
-) -> DeploymentInfo: ...
+) -> DeploymentInfo:
+    ...
 
 
 @inject
diff --git a/src/bentoml/models.py b/src/bentoml/models.py
index 75ad5ccffeb..06006cbf313 100644
--- a/src/bentoml/models.py
+++ b/src/bentoml/models.py
@@ -246,7 +246,8 @@ class CreateKwargs(t.TypedDict):
 @contextmanager
 def _create(
     name: Tag | str, /, **attrs: t.Unpack[CreateKwargs]
-) -> t.Generator[Model, None, None]: ...
+) -> t.Generator[Model, None, None]:
+    ...
 
 
 @t.overload
@@ -264,7 +265,8 @@ def _create(
     metadata: dict[str, t.Any] | None = ...,
     context: ModelContext,
     _model_store: ModelStore = ...,
-) -> t.Generator[Model, None, None]: ...
+) -> t.Generator[Model, None, None]:
+    ...
 
 
 @inject
diff --git a/src/bentoml/triton.py b/src/bentoml/triton.py
index 9968cbf02c7..5e861116786 100644
--- a/src/bentoml/triton.py
+++ b/src/bentoml/triton.py
@@ -203,14 +203,16 @@ def __getattr__(self, item: t.Literal["__attrs_init__"]) -> t.Callable[..., None
     @t.overload
     def __getattr__(
         self, item: _ClientMethod
-    ) -> t.Callable[..., t.Coroutine[t.Any, t.Any, t.Any]]: ...
+    ) -> t.Callable[..., t.Coroutine[t.Any, t.Any, t.Any]]:
+        ...
 
     @t.overload
     def __getattr__(
         self, item: _ModelName
     ) -> _RunnerMethod[
         t.Any, _P, _tritongrpcclient.InferResult | _tritonhttpclient.InferResult
-    ]: ...
+    ]:
+        ...
 
     def __getattr__(self, item: str) -> t.Any:
         from ._internal.runner.runner_handle.remote import TritonRunnerHandle
diff --git a/src/bentoml_cli/utils.py b/src/bentoml_cli/utils.py
index 9321f530ca4..2efce086ad8 100644
--- a/src/bentoml_cli/utils.py
+++ b/src/bentoml_cli/utils.py
@@ -33,7 +33,8 @@ class ClickFunctionWrapper(t.Protocol[P]):
         def __call__(  # pylint: disable=no-method-argument
             *args: P.args,
             **kwargs: P.kwargs,
-        ) -> F[P]: ...
+        ) -> F[P]:
+            ...
 
     WrappedCLI = t.Callable[P, ClickFunctionWrapper[P]]
 
@@ -127,11 +128,13 @@ def validate_docker_tag(
 
 
 @t.overload
-def normalize_none_type(value: t.Mapping[str, t.Any]) -> t.Mapping[str, t.Any]: ...
+def normalize_none_type(value: t.Mapping[str, t.Any]) -> t.Mapping[str, t.Any]:
+    ...
 
 
 @t.overload
-def normalize_none_type(value: ClickParamType) -> ClickParamType: ...
+def normalize_none_type(value: ClickParamType) -> ClickParamType:
+    ...
 
 
 def normalize_none_type(
diff --git a/typings/fs/opener/registry.pyi b/typings/fs/opener/registry.pyi
index bfb53ddd0b5..78786e93da7 100644
--- a/typings/fs/opener/registry.pyi
+++ b/typings/fs/opener/registry.pyi
@@ -11,8 +11,7 @@ from typing import Union
 from ..base import FS
 from .base import Opener
 
-if typing.TYPE_CHECKING:
-    ...
+if typing.TYPE_CHECKING: ...
 
 class Registry:
     def __init__(self, default_opener: Text = ..., load_extern: bool = ...) -> None: ...
diff --git a/typings/fs/permissions.pyi b/typings/fs/permissions.pyi
index c1c7777c265..b91b6a2c9a6 100644
--- a/typings/fs/permissions.pyi
+++ b/typings/fs/permissions.pyi
@@ -8,8 +8,7 @@ from typing import Tuple
 from typing import Type
 from typing import Union
 
-if typing.TYPE_CHECKING:
-    ...
+if typing.TYPE_CHECKING: ...
 
 def make_mode(init: Union[int, Iterable[Text], None]) -> int: ...
 
diff --git a/typings/yaml/representer.pyi b/typings/yaml/representer.pyi
index 0d0a295ae0a..99fb849ee93 100644
--- a/typings/yaml/representer.pyi
+++ b/typings/yaml/representer.pyi
@@ -74,8 +74,7 @@ class SafeRepresenter(BaseRepresenter):
     def represent_none(self, data: NoneType) -> ScalarNode: ...
     def represent_str(self, data: str) -> ScalarNode: ...
 
-    if sys.version_info < (3, 0):
-        ...
+    if sys.version_info < (3, 0): ...
 
     def represent_binary(self, data: bytes) -> ScalarNode: ...
     def represent_bool(self, data: bool) -> ScalarNode: ...
@@ -94,8 +93,7 @@ class SafeRepresenter(BaseRepresenter):
     def represent_undefined(self, data: Any) -> NoReturn: ...
 
 class Representer(SafeRepresenter):
-    if sys.version_info < (3, 0):
-        ...
+    if sys.version_info < (3, 0): ...
 
     def represent_complex(self, data: complex) -> ScalarNode: ...
     def represent_tuple(self, data: Iterable[Any]) -> SequenceNode: ...