modal 1.2.1.dev14__py3-none-any.whl → 1.2.1.dev15__py3-none-any.whl

modal/client.pyi

@@ -33,7 +33,7 @@ class _Client:
         server_url: str,
         client_type: int,
         credentials: typing.Optional[tuple[str, str]],
-        version: str = "1.2.1.dev14",
+        version: str = "1.2.1.dev15",
     ):
         """mdmd:hidden
         The Modal client object is not intended to be instantiated directly by users.
@@ -164,7 +164,7 @@ class Client:
         server_url: str,
         client_type: int,
         credentials: typing.Optional[tuple[str, str]],
-        version: str = "1.2.1.dev14",
+        version: str = "1.2.1.dev15",
     ):
         """mdmd:hidden
         The Modal client object is not intended to be instantiated directly by users.

modal/experimental/flash.py

@@ -321,7 +321,7 @@ class _FlashPrometheusAutoscaler:
 
     async def _compute_target_containers(self, current_replicas: int) -> int:
         """
-        Gets internal metrics from container to autoscale up or down.
+        Gets metrics from container to autoscale up or down.
         """
         containers = await self._get_all_containers()
         if len(containers) > current_replicas:
@@ -334,7 +334,7 @@ class _FlashPrometheusAutoscaler:
         if current_replicas == 0:
             return 1
 
-        # Get metrics based on autoscaler type (prometheus or internal)
+        # Get metrics based on autoscaler type
         sum_metric, n_containers_with_metrics = await self._get_scaling_info(containers)
 
         desired_replicas = self._calculate_desired_replicas(
@@ -406,39 +406,26 @@ class _FlashPrometheusAutoscaler:
         return desired_replicas
 
     async def _get_scaling_info(self, containers) -> tuple[float, int]:
-        """Get metrics using either internal container metrics API or prometheus HTTP endpoints."""
-        if self.metrics_endpoint == "internal":
-            container_metrics_results = await asyncio.gather(
-                *[self._get_container_metrics(container.task_id) for container in containers]
-            )
-            container_metrics_list = []
-            for container_metric in container_metrics_results:
-                if container_metric is None:
-                    continue
-                container_metrics_list.append(getattr(container_metric.metrics, self.target_metric))
-
-            sum_metric = sum(container_metrics_list)
-            n_containers_with_metrics = len(container_metrics_list)
-        else:
-            sum_metric = 0
-            n_containers_with_metrics = 0
-
-            container_metrics_list = await asyncio.gather(
-                *[
-                    self._get_metrics(f"https://{container.host}:{container.port}/{self.metrics_endpoint}")
-                    for container in containers
-                ]
-            )
+        """Get metrics using container exposed metrics endpoints."""
+        sum_metric = 0
+        n_containers_with_metrics = 0
+
+        container_metrics_list = await asyncio.gather(
+            *[
+                self._get_metrics(f"https://{container.host}:{container.port}/{self.metrics_endpoint}")
+                for container in containers
+            ]
+        )
 
-            for container_metrics in container_metrics_list:
-                if (
-                    container_metrics is None
-                    or self.target_metric not in container_metrics
-                    or len(container_metrics[self.target_metric]) == 0
-                ):
-                    continue
-                sum_metric += container_metrics[self.target_metric][0].value
-                n_containers_with_metrics += 1
+        for container_metrics in container_metrics_list:
+            if (
+                container_metrics is None
+                or self.target_metric not in container_metrics
+                or len(container_metrics[self.target_metric]) == 0
+            ):
+                continue
+            sum_metric += container_metrics[self.target_metric][0].value
+            n_containers_with_metrics += 1
 
         return sum_metric, n_containers_with_metrics
 
@@ -474,15 +461,6 @@ class _FlashPrometheusAutoscaler:
 
         return metrics
 
-    async def _get_container_metrics(self, container_id: str) -> Optional[api_pb2.TaskGetAutoscalingMetricsResponse]:
-        req = api_pb2.TaskGetAutoscalingMetricsRequest(task_id=container_id)
-        try:
-            resp = await retry_transient_errors(self.client.stub.TaskGetAutoscalingMetrics, req)
-            return resp
-        except Exception as e:
-            logger.warning(f"[Modal Flash] Error getting metrics for container {container_id}: {e}")
-            return None
-
     async def _get_all_containers(self):
         req = api_pb2.FlashContainerListRequest(function_id=self.fn.object_id)
         resp = await retry_transient_errors(self.client.stub.FlashContainerList, req)
@@ -572,14 +550,10 @@ async def flash_prometheus_autoscaler(
     app_name: str,
     cls_name: str,
     # Endpoint to fetch metrics from. Must be in Prometheus format. Example: "/metrics"
-    # If metrics_endpoint is "internal", we will use containers' internal metrics to autoscale instead.
     metrics_endpoint: str,
     # Target metric to autoscale on. Example: "vllm:num_requests_running"
-    # If metrics_endpoint is "internal", target_metrics options are: [cpu_usage_percent, memory_usage_percent]
     target_metric: str,
     # Target metric value. Example: 25
-    # If metrics_endpoint is "internal", target_metric_value is a percentage value between 0.1 and 1.0 (inclusive),
-    # indicating container's usage of that metric.
     target_metric_value: float,
     min_containers: Optional[int] = None,
     max_containers: Optional[int] = None,

modal/experimental/flash.pyi

@@ -1,5 +1,4 @@
 import modal.client
-import modal_proto.api_pb2
 import subprocess
 import typing
 import typing_extensions
@@ -139,7 +138,7 @@ class _FlashPrometheusAutoscaler:
     async def start(self): ...
     async def _run_autoscaler_loop(self): ...
     async def _compute_target_containers(self, current_replicas: int) -> int:
-        """Gets internal metrics from container to autoscale up or down."""
+        """Gets metrics from container to autoscale up or down."""
         ...
 
     def _calculate_desired_replicas(
@@ -154,13 +153,10 @@ class _FlashPrometheusAutoscaler:
         ...
 
     async def _get_scaling_info(self, containers) -> tuple[float, int]:
-        """Get metrics using either internal container metrics API or prometheus HTTP endpoints."""
+        """Get metrics using container exposed metrics endpoints."""
         ...
 
     async def _get_metrics(self, url: str) -> typing.Optional[dict[str, list[typing.Any]]]: ...
-    async def _get_container_metrics(
-        self, container_id: str
-    ) -> typing.Optional[modal_proto.api_pb2.TaskGetAutoscalingMetricsResponse]: ...
     async def _get_all_containers(self): ...
     async def _set_target_slots(self, target_slots: int): ...
     def _make_scaling_decision(
@@ -226,11 +222,11 @@ class FlashPrometheusAutoscaler:
 
     class ___compute_target_containers_spec(typing_extensions.Protocol[SUPERSELF]):
         def __call__(self, /, current_replicas: int) -> int:
-            """Gets internal metrics from container to autoscale up or down."""
+            """Gets metrics from container to autoscale up or down."""
             ...
 
         async def aio(self, /, current_replicas: int) -> int:
-            """Gets internal metrics from container to autoscale up or down."""
+            """Gets metrics from container to autoscale up or down."""
             ...
 
     _compute_target_containers: ___compute_target_containers_spec[typing_extensions.Self]
@@ -248,11 +244,11 @@ class FlashPrometheusAutoscaler:
 
     class ___get_scaling_info_spec(typing_extensions.Protocol[SUPERSELF]):
         def __call__(self, /, containers) -> tuple[float, int]:
-            """Get metrics using either internal container metrics API or prometheus HTTP endpoints."""
+            """Get metrics using container exposed metrics endpoints."""
             ...
 
         async def aio(self, /, containers) -> tuple[float, int]:
-            """Get metrics using either internal container metrics API or prometheus HTTP endpoints."""
+            """Get metrics using container exposed metrics endpoints."""
             ...
 
     _get_scaling_info: ___get_scaling_info_spec[typing_extensions.Self]
@@ -263,16 +259,6 @@ class FlashPrometheusAutoscaler:
 
     _get_metrics: ___get_metrics_spec[typing_extensions.Self]
 
-    class ___get_container_metrics_spec(typing_extensions.Protocol[SUPERSELF]):
-        def __call__(
-            self, /, container_id: str
-        ) -> typing.Optional[modal_proto.api_pb2.TaskGetAutoscalingMetricsResponse]: ...
-        async def aio(
-            self, /, container_id: str
-        ) -> typing.Optional[modal_proto.api_pb2.TaskGetAutoscalingMetricsResponse]: ...
-
-    _get_container_metrics: ___get_container_metrics_spec[typing_extensions.Self]
-
     class ___get_all_containers_spec(typing_extensions.Protocol[SUPERSELF]):
         def __call__(self, /): ...
         async def aio(self, /): ...

modal/functions.pyi

@@ -401,7 +401,7 @@ class Function(
 
     _call_generator: ___call_generator_spec[typing_extensions.Self]
 
-    class __remote_spec(typing_extensions.Protocol[P_INNER, ReturnType_INNER, SUPERSELF]):
+    class __remote_spec(typing_extensions.Protocol[ReturnType_INNER, P_INNER, SUPERSELF]):
         def __call__(self, /, *args: P_INNER.args, **kwargs: P_INNER.kwargs) -> ReturnType_INNER:
             """Calls the function remotely, executing it with the given arguments and returning the execution's result."""
             ...
@@ -410,7 +410,7 @@ class Function(
             """Calls the function remotely, executing it with the given arguments and returning the execution's result."""
             ...
 
-    remote: __remote_spec[modal._functions.P, modal._functions.ReturnType, typing_extensions.Self]
+    remote: __remote_spec[modal._functions.ReturnType, modal._functions.P, typing_extensions.Self]
 
     class __remote_gen_spec(typing_extensions.Protocol[SUPERSELF]):
         def __call__(self, /, *args, **kwargs) -> typing.Generator[typing.Any, None, None]:
@@ -437,7 +437,7 @@ class Function(
         """
         ...
 
-    class ___experimental_spawn_spec(typing_extensions.Protocol[P_INNER, ReturnType_INNER, SUPERSELF]):
+    class ___experimental_spawn_spec(typing_extensions.Protocol[ReturnType_INNER, P_INNER, SUPERSELF]):
         def __call__(self, /, *args: P_INNER.args, **kwargs: P_INNER.kwargs) -> FunctionCall[ReturnType_INNER]:
             """[Experimental] Calls the function with the given arguments, without waiting for the results.
 
@@ -461,7 +461,7 @@ class Function(
             ...
 
     _experimental_spawn: ___experimental_spawn_spec[
-        modal._functions.P, modal._functions.ReturnType, typing_extensions.Self
+        modal._functions.ReturnType, modal._functions.P, typing_extensions.Self
     ]
 
     class ___spawn_map_inner_spec(typing_extensions.Protocol[P_INNER, SUPERSELF]):
@@ -470,7 +470,7 @@ class Function(
 
     _spawn_map_inner: ___spawn_map_inner_spec[modal._functions.P, typing_extensions.Self]
 
-    class __spawn_spec(typing_extensions.Protocol[P_INNER, ReturnType_INNER, SUPERSELF]):
+    class __spawn_spec(typing_extensions.Protocol[ReturnType_INNER, P_INNER, SUPERSELF]):
         def __call__(self, /, *args: P_INNER.args, **kwargs: P_INNER.kwargs) -> FunctionCall[ReturnType_INNER]:
             """Calls the function with the given arguments, without waiting for the results.
 
@@ -491,7 +491,7 @@ class Function(
             """
             ...
 
-    spawn: __spawn_spec[modal._functions.P, modal._functions.ReturnType, typing_extensions.Self]
+    spawn: __spawn_spec[modal._functions.ReturnType, modal._functions.P, typing_extensions.Self]
 
     def get_raw_f(self) -> collections.abc.Callable[..., typing.Any]:
         """Return the inner Python object wrapped by this Modal Function."""

{modal-1.2.1.dev14.dist-info → modal-1.2.1.dev15.dist-info}/METADATA

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: modal
-Version: 1.2.1.dev14
+Version: 1.2.1.dev15
 Summary: Python client library for Modal
 Author-email: Modal Labs <support@modal.com>
 License: Apache-2.0

{modal-1.2.1.dev14.dist-info → modal-1.2.1.dev15.dist-info}/RECORD

@@ -24,7 +24,7 @@ modal/app.pyi,sha256=AUV5Rp8qQrZJTP2waoKHFY7rYgsXNMYibMcCAQKuSeo,50544
 modal/billing.py,sha256=zmQ3bcCJlwa4KD1IA_QgdWpm1pn13c-7qfy79iEauYI,195
 modal/call_graph.py,sha256=1g2DGcMIJvRy-xKicuf63IVE98gJSnQsr8R_NVMptNc,2581
 modal/client.py,sha256=kyAIVB3Ay-XKJizQ_1ufUFB__EagV0MLmHJpyYyJ7J0,18636
-modal/client.pyi,sha256=2_DiKPDQsg7NASk_5NqJmNb9HYieLChxPmte-GpCPIY,15831
+modal/client.pyi,sha256=UGtqk5bx_bnjPg7uudfmO0QmKws4uCbox7LUe-OzS98,15831
 modal/cloud_bucket_mount.py,sha256=I2GRXYhOWLIz2kJZjXu75jAm9EJkBNcutGc6jR2ReUw,5928
 modal/cloud_bucket_mount.pyi,sha256=VuUOipMIHqFXMkD-3g2bsoqpSxV5qswlFHDOqPQzYAo,7405
 modal/cls.py,sha256=ZxzivE3fNci4-A5uyBYNAzXMXtdqDg3gnYvgbdy5fhg,40384
@@ -41,7 +41,7 @@ modal/file_io.py,sha256=OSKr77TujcXGJW1iikzYiHckLSmv07QBgBHcxxYEkoI,21456
 modal/file_io.pyi,sha256=xtO6Glf_BFwDE7QiQQo24QqcMf_Vv-iz7WojcGVlLBU,15932
 modal/file_pattern_matcher.py,sha256=A_Kdkej6q7YQyhM_2-BvpFmPqJ0oHb54B6yf9VqvPVE,8116
 modal/functions.py,sha256=kcNHvqeGBxPI7Cgd57NIBBghkfbeFJzXO44WW0jSmao,325
-modal/functions.pyi,sha256=CMwApS396tdElFrjnV6RuL2DTCz4C3jYzYoq1y_LPUQ,37988
+modal/functions.pyi,sha256=Z6VuukLrjASAgf0kV9I6c09WvP_b2gCujX6f9j2bBaw,37988
 modal/gpu.py,sha256=Fe5ORvVPDIstSq1xjmM6OoNgLYFWvogP9r5BgmD3hYg,6769
 modal/image.py,sha256=HDkOnhIAN8g63a8LTN4J5SjC9ciReFQQJIxTS2z5KFM,107216
 modal/image.pyi,sha256=dMvMwAuvWkNN2BRYJFijkEy2m_xtEXgCKK0T7FVldsc,77514
@@ -153,10 +153,10 @@ modal/cli/programs/run_jupyter.py,sha256=IJw8nds8Cjl9j4dxBqMGxhz-bIyVX0kle7jbt8H
 modal/cli/programs/run_marimo.py,sha256=QlCGPkwQ0XLajt2LtujR_BGsRV1AZ3OCUgZxkkM1lug,2893
 modal/cli/programs/vscode.py,sha256=E1aJPU7b8RWWj-JX71DifsCoLYsCAnre7lATiBJjUms,3386
 modal/experimental/__init__.py,sha256=9gkVuDmu3m4TlKoU3MzEtTOemUSs8EEOWba40s7Aa0M,14043
-modal/experimental/flash.py,sha256=C4sef08rARYFllsgtqukFmYL18SZW0_JpMS0BejDcUs,28552
-modal/experimental/flash.pyi,sha256=vV_OQhtdrPn8SW0XrBK-aLLHHIvxAzLzwFbWrke-m74,15463
+modal/experimental/flash.py,sha256=-lSyFBbeT6UT-uB29L955SNh6L6ISg_uBDy5gF4ZpLo,26919
+modal/experimental/flash.pyi,sha256=uwinKAYxpunNNfBj58FP88DXb535Qik4F6tnJKPAIwQ,14696
 modal/experimental/ipython.py,sha256=TrCfmol9LGsRZMeDoeMPx3Hv3BFqQhYnmD_iH0pqdhk,2904
-modal-1.2.1.dev14.dist-info/licenses/LICENSE,sha256=psuoW8kuDP96RQsdhzwOqi6fyWv0ct8CR6Jr7He_P_k,10173
+modal-1.2.1.dev15.dist-info/licenses/LICENSE,sha256=psuoW8kuDP96RQsdhzwOqi6fyWv0ct8CR6Jr7He_P_k,10173
 modal_docs/__init__.py,sha256=svYKtV8HDwDCN86zbdWqyq5T8sMdGDj0PVlzc2tIxDM,28
 modal_docs/gen_cli_docs.py,sha256=c1yfBS_x--gL5bs0N4ihMwqwX8l3IBWSkBAKNNIi6bQ,3801
 modal_docs/gen_reference_docs.py,sha256=d_CQUGQ0rfw28u75I2mov9AlS773z9rG40-yq5o7g2U,6359
@@ -164,13 +164,13 @@ modal_docs/mdmd/__init__.py,sha256=svYKtV8HDwDCN86zbdWqyq5T8sMdGDj0PVlzc2tIxDM,2
 modal_docs/mdmd/mdmd.py,sha256=tUTImNd4UMFk1opkaw8J672gX8AkBO5gbY2S_NMxsxs,7140
 modal_docs/mdmd/signatures.py,sha256=XJaZrK7Mdepk5fdX51A8uENiLFNil85Ud0d4MH8H5f0,3218
 modal_proto/__init__.py,sha256=MIEP8jhXUeGq_eCjYFcqN5b1bxBM4fdk0VESpjWR0fc,28
-modal_proto/api.proto,sha256=yYsIf0_6hfT98xBHfad82fBDAfW7Wcip5g74MZMNNwE,108935
-modal_proto/api_grpc.py,sha256=JYmHBCRazCqwFsSpNMK2kKk8UCGVXi2E9XHJvT6NpSI,135300
-modal_proto/api_pb2.py,sha256=h31c6JqkcvmuFVebu5oEa2FbZTlbOSCanTZdD0yeubU,381165
-modal_proto/api_pb2.pyi,sha256=w8xScmbhxUDCV6yTk9baT6u9h7WLyeTktFqASoyHSb0,532300
-modal_proto/api_pb2_grpc.py,sha256=AIj7DOISi5pKqmXm9APOyrFSW1-8-V5peGebtnMHCeo,291495
-modal_proto/api_pb2_grpc.pyi,sha256=1TXj_sL24TSMit0wL5ScIytnHoD5RI0DT5O1vqOLLYM,68293
-modal_proto/modal_api_grpc.py,sha256=qbz7Ssx_kvfy28aeFLLXLFdoS5dY4OHVrgLcS5rDXKc,20388
+modal_proto/api.proto,sha256=GYn-HjF_vc7Y4_BvHZBCG-c9ZJU0CWnyX_Lk973J2Pw,108645
+modal_proto/api_grpc.py,sha256=vwC-GjejDKWbG5jRN3rkU8WBSqQ8Pdj-T2E2xAECAUw,134411
+modal_proto/api_pb2.py,sha256=EtmyiWUznFw_YzPhrw9U4SuXuDsAgZigozbPR4n4TG4,379625
+modal_proto/api_pb2.pyi,sha256=77N3MpNhleBngbyMkph2w87IZmj9DVxhJ1Ezc1Aosmw,531180
+modal_proto/api_pb2_grpc.py,sha256=Hqw9jcbhpr-W6jsfog_tGU55ouZjITxGvA-DGNBqOLA,289714
+modal_proto/api_pb2_grpc.pyi,sha256=QLJ58ANCx147HeGJva58h0MTCLIDs9JmVjrx8bDdwlg,67776
+modal_proto/modal_api_grpc.py,sha256=ME4PoI4cm7_Tpi3duU6-0O5LwhbYKrgJ5n3ijWgeNss,20256
 modal_proto/py.typed,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
 modal_proto/sandbox_router.proto,sha256=o6LZWekz2h7uAhBlHcu_EqXnc5ptxm-4j_ZUFVGAJFk,5161
 modal_proto/sandbox_router_grpc.py,sha256=27daOTX2N5hADDG-5Qnn4Yj3VfekyJwrDUkrQ12mPuU,5004
@@ -184,10 +184,10 @@ modal_proto/task_command_router_pb2.py,sha256=_pD2ZpU0bNzhwBdzmLoLyLtAtftI_Agxwn
 modal_proto/task_command_router_pb2.pyi,sha256=EyDgXPLr7alqjXYERV8w_MPuO404x0uCppmSkrfE9IE,14589
 modal_proto/task_command_router_pb2_grpc.py,sha256=uEQ0HdrCp8v-9bB5yIic9muA8spCShLHY6Bz9cCgOUE,10114
 modal_proto/task_command_router_pb2_grpc.pyi,sha256=s3Yxsrawdj4nr8vqQqsAxyX6ilWaGbdECy425KKbLIA,3301
-modal_version/__init__.py,sha256=743KmmOlJb-M5EkswhYesVjee8eiLrK4PizfYr2lZCM,121
+modal_version/__init__.py,sha256=vkq2oCnafaqr6nqiR7nHb8JclOJLXYMcCph81nNMr3o,121
 modal_version/__main__.py,sha256=2FO0yYQQwDTh6udt1h-cBnGd1c4ZyHnHSI4BksxzVac,105
-modal-1.2.1.dev14.dist-info/METADATA,sha256=kp21p_YTNC2q_oQRDT9xdagRNkhSrnlIfh-uQayaTqo,2484
-modal-1.2.1.dev14.dist-info/WHEEL,sha256=1tXe9gY0PYatrMPMDd6jXqjfpz_B-Wqm32CPfRC58XU,91
-modal-1.2.1.dev14.dist-info/entry_points.txt,sha256=An-wYgeEUnm6xzrAP9_NTSTSciYvvEWsMZILtYrvpAI,46
-modal-1.2.1.dev14.dist-info/top_level.txt,sha256=4BWzoKYREKUZ5iyPzZpjqx4G8uB5TWxXPDwibLcVa7k,43
-modal-1.2.1.dev14.dist-info/RECORD,,
+modal-1.2.1.dev15.dist-info/METADATA,sha256=er7pk05y0NA4kL_o8Jd0mYSzjqNc54yCFsz_vXY60u4,2484
+modal-1.2.1.dev15.dist-info/WHEEL,sha256=1tXe9gY0PYatrMPMDd6jXqjfpz_B-Wqm32CPfRC58XU,91
+modal-1.2.1.dev15.dist-info/entry_points.txt,sha256=An-wYgeEUnm6xzrAP9_NTSTSciYvvEWsMZILtYrvpAI,46
+modal-1.2.1.dev15.dist-info/top_level.txt,sha256=4BWzoKYREKUZ5iyPzZpjqx4G8uB5TWxXPDwibLcVa7k,43
+modal-1.2.1.dev15.dist-info/RECORD,,

modal_proto/api.proto

@@ -3143,14 +3143,6 @@ message TaskCurrentInputsResponse {
   repeated string input_ids = 1;
 }
 
-message TaskGetAutoscalingMetricsRequest {
-  string task_id = 1;
-}
-
-message TaskGetAutoscalingMetricsResponse {
-  AutoscalingMetrics metrics = 1;
-}
-
 // Used to get a JWT and URL for direct access to a task command router
 // running on the modal-worker, so the client can issue exec commands (and other
 // operations as they become available) directly to the worker.
@@ -3778,7 +3770,6 @@ service ModalClient {
   // Tasks
   rpc TaskClusterHello(TaskClusterHelloRequest) returns (TaskClusterHelloResponse);
   rpc TaskCurrentInputs(google.protobuf.Empty) returns (TaskCurrentInputsResponse);
-  rpc TaskGetAutoscalingMetrics(TaskGetAutoscalingMetricsRequest) returns (TaskGetAutoscalingMetricsResponse); // Used for flash autoscaling
   rpc TaskGetCommandRouterAccess(TaskGetCommandRouterAccessRequest) returns (TaskGetCommandRouterAccessResponse);
   rpc TaskList(TaskListRequest) returns (TaskListResponse);
   rpc TaskResult(TaskResultRequest) returns (google.protobuf.Empty);

modal_proto/api_grpc.py

@@ -610,10 +610,6 @@ class ModalClientBase(abc.ABC):
     async def TaskCurrentInputs(self, stream: 'grpclib.server.Stream[google.protobuf.empty_pb2.Empty, modal_proto.api_pb2.TaskCurrentInputsResponse]') -> None:
         pass
 
-    @abc.abstractmethod
-    async def TaskGetAutoscalingMetrics(self, stream: 'grpclib.server.Stream[modal_proto.api_pb2.TaskGetAutoscalingMetricsRequest, modal_proto.api_pb2.TaskGetAutoscalingMetricsResponse]') -> None:
-        pass
-
     @abc.abstractmethod
     async def TaskGetCommandRouterAccess(self, stream: 'grpclib.server.Stream[modal_proto.api_pb2.TaskGetCommandRouterAccessRequest, modal_proto.api_pb2.TaskGetCommandRouterAccessResponse]') -> None:
         pass
@@ -1608,12 +1604,6 @@ class ModalClientBase(abc.ABC):
                 google.protobuf.empty_pb2.Empty,
                 modal_proto.api_pb2.TaskCurrentInputsResponse,
             ),
-            '/modal.client.ModalClient/TaskGetAutoscalingMetrics': grpclib.const.Handler(
-                self.TaskGetAutoscalingMetrics,
-                grpclib.const.Cardinality.UNARY_UNARY,
-                modal_proto.api_pb2.TaskGetAutoscalingMetricsRequest,
-                modal_proto.api_pb2.TaskGetAutoscalingMetricsResponse,
-            ),
             '/modal.client.ModalClient/TaskGetCommandRouterAccess': grpclib.const.Handler(
                 self.TaskGetCommandRouterAccess,
                 grpclib.const.Cardinality.UNARY_UNARY,
@@ -2664,12 +2654,6 @@ class ModalClientStub:
             google.protobuf.empty_pb2.Empty,
             modal_proto.api_pb2.TaskCurrentInputsResponse,
         )
-        self.TaskGetAutoscalingMetrics = grpclib.client.UnaryUnaryMethod(
-            channel,
-            '/modal.client.ModalClient/TaskGetAutoscalingMetrics',
-            modal_proto.api_pb2.TaskGetAutoscalingMetricsRequest,
-            modal_proto.api_pb2.TaskGetAutoscalingMetricsResponse,
-        )
         self.TaskGetCommandRouterAccess = grpclib.client.UnaryUnaryMethod(
             channel,
             '/modal.client.ModalClient/TaskGetCommandRouterAccess',