databricks-sdk 0.50.0__py3-none-any.whl → 0.51.0__py3-none-any.whl

This diff represents the content of publicly available package versions that have been released to one of the supported registries. The information contained in this diff is provided for informational purposes only and reflects changes between package versions as they appear in their respective public registries.

Potentially problematic release.


This version of databricks-sdk might be problematic. Click here for more details.

@@ -576,10 +576,11 @@ class CleanRoomTaskRunState:
576
576
 
577
577
  life_cycle_state: Optional[CleanRoomTaskRunLifeCycleState] = None
578
578
  """A value indicating the run's current lifecycle state. This field is always available in the
579
- response."""
579
+ response. Note: Additional states might be introduced in future releases."""
580
580
 
581
581
  result_state: Optional[CleanRoomTaskRunResultState] = None
582
- """A value indicating the run's result. This field is only available for terminal lifecycle states."""
582
+ """A value indicating the run's result. This field is only available for terminal lifecycle states.
583
+ Note: Additional states might be introduced in future releases."""
583
584
 
584
585
  def as_dict(self) -> dict:
585
586
  """Serializes the CleanRoomTaskRunState into a dictionary suitable for use as a JSON request body."""
@@ -1023,8 +1024,8 @@ class CreateJob:
1023
1024
  """Job-level parameter definitions"""
1024
1025
 
1025
1026
  performance_target: Optional[PerformanceTarget] = None
1026
- """The performance mode on a serverless job. The performance target determines the level of compute
1027
- performance or cost-efficiency for the run.
1027
+ """The performance mode on a serverless job. This field determines the level of compute performance
1028
+ or cost-efficiency for the run.
1028
1029
 
1029
1030
  * `STANDARD`: Enables cost-efficient execution of serverless workloads. *
1030
1031
  `PERFORMANCE_OPTIMIZED`: Prioritizes fast startup and execution times through rapid scaling and
@@ -1322,11 +1323,14 @@ class DashboardTask:
1322
1323
  """Configures the Lakeview Dashboard job task type."""
1323
1324
 
1324
1325
  dashboard_id: Optional[str] = None
1326
+ """The identifier of the dashboard to refresh."""
1325
1327
 
1326
1328
  subscription: Optional[Subscription] = None
1329
+ """Optional: subscription configuration for sending the dashboard snapshot."""
1327
1330
 
1328
1331
  warehouse_id: Optional[str] = None
1329
- """The warehouse id to execute the dashboard with for the schedule"""
1332
+ """Optional: The warehouse id to execute the dashboard with for the schedule. If not specified, the
1333
+ default warehouse of the dashboard will be used."""
1330
1334
 
1331
1335
  def as_dict(self) -> dict:
1332
1336
  """Serializes the DashboardTask into a dictionary suitable for use as a JSON request body."""
@@ -2703,8 +2707,11 @@ class JobEnvironment:
2703
2707
  """The key of an environment. It has to be unique within a job."""
2704
2708
 
2705
2709
  spec: Optional[compute.Environment] = None
2706
- """The environment entity used to preserve serverless environment side panel and jobs' environment
2707
- for non-notebook task. In this minimal environment spec, only pip dependencies are supported."""
2710
+ """The environment entity used to preserve serverless environment side panel, jobs' environment for
2711
+ non-notebook task, and DLT's environment for classic and serverless pipelines. (Note: DLT uses a
2712
+ copied version of the Environment proto below, at
2713
+ //spark/pipelines/api/protos/copied/libraries-environments-copy.proto) In this minimal
2714
+ environment spec, only pip dependencies are supported."""
2708
2715
 
2709
2716
  def as_dict(self) -> dict:
2710
2717
  """Serializes the JobEnvironment into a dictionary suitable for use as a JSON request body."""
@@ -3111,8 +3118,8 @@ class JobSettings:
3111
3118
  """Job-level parameter definitions"""
3112
3119
 
3113
3120
  performance_target: Optional[PerformanceTarget] = None
3114
- """The performance mode on a serverless job. The performance target determines the level of compute
3115
- performance or cost-efficiency for the run.
3121
+ """The performance mode on a serverless job. This field determines the level of compute performance
3122
+ or cost-efficiency for the run.
3116
3123
 
3117
3124
  * `STANDARD`: Enables cost-efficient execution of serverless workloads. *
3118
3125
  `PERFORMANCE_OPTIMIZED`: Prioritizes fast startup and execution times through rapid scaling and
@@ -4174,6 +4181,15 @@ class QueueSettings:
4174
4181
 
4175
4182
  @dataclass
4176
4183
  class RepairHistoryItem:
4184
+ effective_performance_target: Optional[PerformanceTarget] = None
4185
+ """The actual performance target used by the serverless run during execution. This can differ from
4186
+ the client-set performance target on the request depending on whether the performance mode is
4187
+ supported by the job type.
4188
+
4189
+ * `STANDARD`: Enables cost-efficient execution of serverless workloads. *
4190
+ `PERFORMANCE_OPTIMIZED`: Prioritizes fast startup and execution times through rapid scaling and
4191
+ optimized cluster performance."""
4192
+
4177
4193
  end_time: Optional[int] = None
4178
4194
  """The end time of the (repaired) run."""
4179
4195
 
@@ -4198,6 +4214,8 @@ class RepairHistoryItem:
4198
4214
  def as_dict(self) -> dict:
4199
4215
  """Serializes the RepairHistoryItem into a dictionary suitable for use as a JSON request body."""
4200
4216
  body = {}
4217
+ if self.effective_performance_target is not None:
4218
+ body["effective_performance_target"] = self.effective_performance_target.value
4201
4219
  if self.end_time is not None:
4202
4220
  body["end_time"] = self.end_time
4203
4221
  if self.id is not None:
@@ -4217,6 +4235,8 @@ class RepairHistoryItem:
4217
4235
  def as_shallow_dict(self) -> dict:
4218
4236
  """Serializes the RepairHistoryItem into a shallow dictionary of its immediate attributes."""
4219
4237
  body = {}
4238
+ if self.effective_performance_target is not None:
4239
+ body["effective_performance_target"] = self.effective_performance_target
4220
4240
  if self.end_time is not None:
4221
4241
  body["end_time"] = self.end_time
4222
4242
  if self.id is not None:
@@ -4237,6 +4257,7 @@ class RepairHistoryItem:
4237
4257
  def from_dict(cls, d: Dict[str, Any]) -> RepairHistoryItem:
4238
4258
  """Deserializes the RepairHistoryItem from a dictionary."""
4239
4259
  return cls(
4260
+ effective_performance_target=_enum(d, "effective_performance_target", PerformanceTarget),
4240
4261
  end_time=d.get("end_time", None),
4241
4262
  id=d.get("id", None),
4242
4263
  start_time=d.get("start_time", None),
@@ -4298,6 +4319,15 @@ class RepairRun:
4298
4319
  [Task parameter variables]: https://docs.databricks.com/jobs.html#parameter-variables
4299
4320
  [dbutils.widgets.get]: https://docs.databricks.com/dev-tools/databricks-utils.html"""
4300
4321
 
4322
+ performance_target: Optional[PerformanceTarget] = None
4323
+ """The performance mode on a serverless job. The performance target determines the level of compute
4324
+ performance or cost-efficiency for the run. This field overrides the performance target defined
4325
+ on the job level.
4326
+
4327
+ * `STANDARD`: Enables cost-efficient execution of serverless workloads. *
4328
+ `PERFORMANCE_OPTIMIZED`: Prioritizes fast startup and execution times through rapid scaling and
4329
+ optimized cluster performance."""
4330
+
4301
4331
  pipeline_params: Optional[PipelineParams] = None
4302
4332
  """Controls whether the pipeline should perform a full refresh"""
4303
4333
 
@@ -4364,6 +4394,8 @@ class RepairRun:
4364
4394
  body["latest_repair_id"] = self.latest_repair_id
4365
4395
  if self.notebook_params:
4366
4396
  body["notebook_params"] = self.notebook_params
4397
+ if self.performance_target is not None:
4398
+ body["performance_target"] = self.performance_target.value
4367
4399
  if self.pipeline_params:
4368
4400
  body["pipeline_params"] = self.pipeline_params.as_dict()
4369
4401
  if self.python_named_params:
@@ -4397,6 +4429,8 @@ class RepairRun:
4397
4429
  body["latest_repair_id"] = self.latest_repair_id
4398
4430
  if self.notebook_params:
4399
4431
  body["notebook_params"] = self.notebook_params
4432
+ if self.performance_target is not None:
4433
+ body["performance_target"] = self.performance_target
4400
4434
  if self.pipeline_params:
4401
4435
  body["pipeline_params"] = self.pipeline_params
4402
4436
  if self.python_named_params:
@@ -4426,6 +4460,7 @@ class RepairRun:
4426
4460
  job_parameters=d.get("job_parameters", None),
4427
4461
  latest_repair_id=d.get("latest_repair_id", None),
4428
4462
  notebook_params=d.get("notebook_params", None),
4463
+ performance_target=_enum(d, "performance_target", PerformanceTarget),
4429
4464
  pipeline_params=_from_dict(d, "pipeline_params", PipelineParams),
4430
4465
  python_named_params=d.get("python_named_params", None),
4431
4466
  python_params=d.get("python_params", None),
@@ -6020,13 +6055,14 @@ class RunState:
6020
6055
 
6021
6056
  life_cycle_state: Optional[RunLifeCycleState] = None
6022
6057
  """A value indicating the run's current lifecycle state. This field is always available in the
6023
- response."""
6058
+ response. Note: Additional states might be introduced in future releases."""
6024
6059
 
6025
6060
  queue_reason: Optional[str] = None
6026
6061
  """The reason indicating why the run was queued."""
6027
6062
 
6028
6063
  result_state: Optional[RunResultState] = None
6029
- """A value indicating the run's result. This field is only available for terminal lifecycle states."""
6064
+ """A value indicating the run's result. This field is only available for terminal lifecycle states.
6065
+ Note: Additional states might be introduced in future releases."""
6030
6066
 
6031
6067
  state_message: Optional[str] = None
6032
6068
  """A descriptive message for the current state. This field is unstructured, and its exact format is
@@ -6161,7 +6197,7 @@ class RunTask:
6161
6197
  does not support retries or notifications."""
6162
6198
 
6163
6199
  dashboard_task: Optional[DashboardTask] = None
6164
- """The task runs a DashboardTask when the `dashboard_task` field is present."""
6200
+ """The task refreshes a dashboard and sends a snapshot to subscribers."""
6165
6201
 
6166
6202
  dbt_task: Optional[DbtTask] = None
6167
6203
  """The task runs one or more dbt commands when the `dbt_task` field is present. The dbt task
@@ -7549,7 +7585,7 @@ class SubmitTask:
7549
7585
  does not support retries or notifications."""
7550
7586
 
7551
7587
  dashboard_task: Optional[DashboardTask] = None
7552
- """The task runs a DashboardTask when the `dashboard_task` field is present."""
7588
+ """The task refreshes a dashboard and sends a snapshot to subscribers."""
7553
7589
 
7554
7590
  dbt_task: Optional[DbtTask] = None
7555
7591
  """The task runs one or more dbt commands when the `dbt_task` field is present. The dbt task
@@ -7818,6 +7854,7 @@ class Subscription:
7818
7854
  """When true, the subscription will not send emails."""
7819
7855
 
7820
7856
  subscribers: Optional[List[SubscriptionSubscriber]] = None
7857
+ """The list of subscribers to send the snapshot of the dashboard to."""
7821
7858
 
7822
7859
  def as_dict(self) -> dict:
7823
7860
  """Serializes the Subscription into a dictionary suitable for use as a JSON request body."""
@@ -7854,8 +7891,12 @@ class Subscription:
7854
7891
  @dataclass
7855
7892
  class SubscriptionSubscriber:
7856
7893
  destination_id: Optional[str] = None
7894
+ """A snapshot of the dashboard will be sent to the destination when the `destination_id` field is
7895
+ present."""
7857
7896
 
7858
7897
  user_name: Optional[str] = None
7898
+ """A snapshot of the dashboard will be sent to the user's email when the `user_name` field is
7899
+ present."""
7859
7900
 
7860
7901
  def as_dict(self) -> dict:
7861
7902
  """Serializes the SubscriptionSubscriber into a dictionary suitable for use as a JSON request body."""
@@ -7954,7 +7995,7 @@ class Task:
7954
7995
  does not support retries or notifications."""
7955
7996
 
7956
7997
  dashboard_task: Optional[DashboardTask] = None
7957
- """The task runs a DashboardTask when the `dashboard_task` field is present."""
7998
+ """The task refreshes a dashboard and sends a snapshot to subscribers."""
7958
7999
 
7959
8000
  dbt_task: Optional[DbtTask] = None
7960
8001
  """The task runs one or more dbt commands when the `dbt_task` field is present. The dbt task
@@ -8459,7 +8500,7 @@ class TerminationCodeCode(Enum):
8459
8500
  invalid configuration. Refer to the state message for further details. * `CLOUD_FAILURE`: The
8460
8501
  run failed due to a cloud provider issue. Refer to the state message for further details. *
8461
8502
  `MAX_JOB_QUEUE_SIZE_EXCEEDED`: The run was skipped due to reaching the job level queue size
8462
- limit.
8503
+ limit. * `DISABLED`: The run was never executed because it was disabled explicitly by the user.
8463
8504
 
8464
8505
  [Link]: https://kb.databricks.com/en_US/notebooks/too-many-execution-contexts-are-open-right-now"""
8465
8506
 
@@ -8468,6 +8509,7 @@ class TerminationCodeCode(Enum):
8468
8509
  CLOUD_FAILURE = "CLOUD_FAILURE"
8469
8510
  CLUSTER_ERROR = "CLUSTER_ERROR"
8470
8511
  CLUSTER_REQUEST_LIMIT_EXCEEDED = "CLUSTER_REQUEST_LIMIT_EXCEEDED"
8512
+ DISABLED = "DISABLED"
8471
8513
  DRIVER_ERROR = "DRIVER_ERROR"
8472
8514
  FEATURE_DISABLED = "FEATURE_DISABLED"
8473
8515
  INTERNAL_ERROR = "INTERNAL_ERROR"
@@ -8523,7 +8565,7 @@ class TerminationDetails:
8523
8565
  invalid configuration. Refer to the state message for further details. * `CLOUD_FAILURE`: The
8524
8566
  run failed due to a cloud provider issue. Refer to the state message for further details. *
8525
8567
  `MAX_JOB_QUEUE_SIZE_EXCEEDED`: The run was skipped due to reaching the job level queue size
8526
- limit.
8568
+ limit. * `DISABLED`: The run was never executed because it was disabled explicitly by the user.
8527
8569
 
8528
8570
  [Link]: https://kb.databricks.com/en_US/notebooks/too-many-execution-contexts-are-open-right-now"""
8529
8571
 
@@ -9140,8 +9182,8 @@ class JobsAPI:
9140
9182
  :param parameters: List[:class:`JobParameterDefinition`] (optional)
9141
9183
  Job-level parameter definitions
9142
9184
  :param performance_target: :class:`PerformanceTarget` (optional)
9143
- The performance mode on a serverless job. The performance target determines the level of compute
9144
- performance or cost-efficiency for the run.
9185
+ The performance mode on a serverless job. This field determines the level of compute performance or
9186
+ cost-efficiency for the run.
9145
9187
 
9146
9188
  * `STANDARD`: Enables cost-efficient execution of serverless workloads. * `PERFORMANCE_OPTIMIZED`:
9147
9189
  Prioritizes fast startup and execution times through rapid scaling and optimized cluster
@@ -9593,6 +9635,7 @@ class JobsAPI:
9593
9635
  job_parameters: Optional[Dict[str, str]] = None,
9594
9636
  latest_repair_id: Optional[int] = None,
9595
9637
  notebook_params: Optional[Dict[str, str]] = None,
9638
+ performance_target: Optional[PerformanceTarget] = None,
9596
9639
  pipeline_params: Optional[PipelineParams] = None,
9597
9640
  python_named_params: Optional[Dict[str, str]] = None,
9598
9641
  python_params: Optional[List[str]] = None,
@@ -9643,6 +9686,14 @@ class JobsAPI:
9643
9686
 
9644
9687
  [Task parameter variables]: https://docs.databricks.com/jobs.html#parameter-variables
9645
9688
  [dbutils.widgets.get]: https://docs.databricks.com/dev-tools/databricks-utils.html
9689
+ :param performance_target: :class:`PerformanceTarget` (optional)
9690
+ The performance mode on a serverless job. The performance target determines the level of compute
9691
+ performance or cost-efficiency for the run. This field overrides the performance target defined on
9692
+ the job level.
9693
+
9694
+ * `STANDARD`: Enables cost-efficient execution of serverless workloads. * `PERFORMANCE_OPTIMIZED`:
9695
+ Prioritizes fast startup and execution times through rapid scaling and optimized cluster
9696
+ performance.
9646
9697
  :param pipeline_params: :class:`PipelineParams` (optional)
9647
9698
  Controls whether the pipeline should perform a full refresh
9648
9699
  :param python_named_params: Dict[str,str] (optional)
@@ -9703,6 +9754,8 @@ class JobsAPI:
9703
9754
  body["latest_repair_id"] = latest_repair_id
9704
9755
  if notebook_params is not None:
9705
9756
  body["notebook_params"] = notebook_params
9757
+ if performance_target is not None:
9758
+ body["performance_target"] = performance_target.value
9706
9759
  if pipeline_params is not None:
9707
9760
  body["pipeline_params"] = pipeline_params.as_dict()
9708
9761
  if python_named_params is not None:
@@ -9742,6 +9795,7 @@ class JobsAPI:
9742
9795
  job_parameters: Optional[Dict[str, str]] = None,
9743
9796
  latest_repair_id: Optional[int] = None,
9744
9797
  notebook_params: Optional[Dict[str, str]] = None,
9798
+ performance_target: Optional[PerformanceTarget] = None,
9745
9799
  pipeline_params: Optional[PipelineParams] = None,
9746
9800
  python_named_params: Optional[Dict[str, str]] = None,
9747
9801
  python_params: Optional[List[str]] = None,
@@ -9758,6 +9812,7 @@ class JobsAPI:
9758
9812
  job_parameters=job_parameters,
9759
9813
  latest_repair_id=latest_repair_id,
9760
9814
  notebook_params=notebook_params,
9815
+ performance_target=performance_target,
9761
9816
  pipeline_params=pipeline_params,
9762
9817
  python_named_params=python_named_params,
9763
9818
  python_params=python_params,
@@ -1191,10 +1191,10 @@ class AccountFederationPolicyAPI:
1191
1191
  def __init__(self, api_client):
1192
1192
  self._api = api_client
1193
1193
 
1194
- def create(self, *, policy: Optional[FederationPolicy] = None, policy_id: Optional[str] = None) -> FederationPolicy:
1194
+ def create(self, policy: FederationPolicy, *, policy_id: Optional[str] = None) -> FederationPolicy:
1195
1195
  """Create account federation policy.
1196
1196
 
1197
- :param policy: :class:`FederationPolicy` (optional)
1197
+ :param policy: :class:`FederationPolicy`
1198
1198
  :param policy_id: str (optional)
1199
1199
  The identifier for the federation policy. The identifier must contain only lowercase alphanumeric
1200
1200
  characters, numbers, hyphens, and slashes. If unspecified, the id will be assigned by Databricks.
@@ -1284,13 +1284,13 @@ class AccountFederationPolicyAPI:
1284
1284
  query["page_token"] = json["next_page_token"]
1285
1285
 
1286
1286
  def update(
1287
- self, policy_id: str, *, policy: Optional[FederationPolicy] = None, update_mask: Optional[str] = None
1287
+ self, policy_id: str, policy: FederationPolicy, *, update_mask: Optional[str] = None
1288
1288
  ) -> FederationPolicy:
1289
1289
  """Update account federation policy.
1290
1290
 
1291
1291
  :param policy_id: str
1292
1292
  The identifier for the federation policy.
1293
- :param policy: :class:`FederationPolicy` (optional)
1293
+ :param policy: :class:`FederationPolicy`
1294
1294
  :param update_mask: str (optional)
1295
1295
  The field mask specifies which fields of the policy to update. To specify multiple fields in the
1296
1296
  field mask, use comma as the separator (no space). The special value '*' indicates that all fields
@@ -1758,13 +1758,13 @@ class ServicePrincipalFederationPolicyAPI:
1758
1758
  self._api = api_client
1759
1759
 
1760
1760
  def create(
1761
- self, service_principal_id: int, *, policy: Optional[FederationPolicy] = None, policy_id: Optional[str] = None
1761
+ self, service_principal_id: int, policy: FederationPolicy, *, policy_id: Optional[str] = None
1762
1762
  ) -> FederationPolicy:
1763
1763
  """Create service principal federation policy.
1764
1764
 
1765
1765
  :param service_principal_id: int
1766
1766
  The service principal id for the federation policy.
1767
- :param policy: :class:`FederationPolicy` (optional)
1767
+ :param policy: :class:`FederationPolicy`
1768
1768
  :param policy_id: str (optional)
1769
1769
  The identifier for the federation policy. The identifier must contain only lowercase alphanumeric
1770
1770
  characters, numbers, hyphens, and slashes. If unspecified, the id will be assigned by Databricks.
@@ -1869,12 +1869,7 @@ class ServicePrincipalFederationPolicyAPI:
1869
1869
  query["page_token"] = json["next_page_token"]
1870
1870
 
1871
1871
  def update(
1872
- self,
1873
- service_principal_id: int,
1874
- policy_id: str,
1875
- *,
1876
- policy: Optional[FederationPolicy] = None,
1877
- update_mask: Optional[str] = None,
1872
+ self, service_principal_id: int, policy_id: str, policy: FederationPolicy, *, update_mask: Optional[str] = None
1878
1873
  ) -> FederationPolicy:
1879
1874
  """Update service principal federation policy.
1880
1875
 
@@ -1882,7 +1877,7 @@ class ServicePrincipalFederationPolicyAPI:
1882
1877
  The service principal id for the federation policy.
1883
1878
  :param policy_id: str
1884
1879
  The identifier for the federation policy.
1885
- :param policy: :class:`FederationPolicy` (optional)
1880
+ :param policy: :class:`FederationPolicy`
1886
1881
  :param update_mask: str (optional)
1887
1882
  The field mask specifies which fields of the policy to update. To specify multiple fields in the
1888
1883
  field mask, use comma as the separator (no space). The special value '*' indicates that all fields
@@ -2874,7 +2874,8 @@ class ServedEntityInput:
2874
2874
  """The workload size of the served entity. The workload size corresponds to a range of provisioned
2875
2875
  concurrency that the compute autoscales between. A single unit of provisioned concurrency can
2876
2876
  process one request at a time. Valid workload sizes are "Small" (4 - 4 provisioned concurrency),
2877
- "Medium" (8 - 16 provisioned concurrency), and "Large" (16 - 64 provisioned concurrency). If
2877
+ "Medium" (8 - 16 provisioned concurrency), and "Large" (16 - 64 provisioned concurrency).
2878
+ Additional custom workload sizes can also be used when available in the workspace. If
2878
2879
  scale-to-zero is enabled, the lower bound of the provisioned concurrency for each workload size
2879
2880
  is 0."""
2880
2881
 
@@ -3014,7 +3015,8 @@ class ServedEntityOutput:
3014
3015
  """The workload size of the served entity. The workload size corresponds to a range of provisioned
3015
3016
  concurrency that the compute autoscales between. A single unit of provisioned concurrency can
3016
3017
  process one request at a time. Valid workload sizes are "Small" (4 - 4 provisioned concurrency),
3017
- "Medium" (8 - 16 provisioned concurrency), and "Large" (16 - 64 provisioned concurrency). If
3018
+ "Medium" (8 - 16 provisioned concurrency), and "Large" (16 - 64 provisioned concurrency).
3019
+ Additional custom workload sizes can also be used when available in the workspace. If
3018
3020
  scale-to-zero is enabled, the lower bound of the provisioned concurrency for each workload size
3019
3021
  is 0."""
3020
3022
 
@@ -3204,11 +3206,12 @@ class ServedModelInput:
3204
3206
  model, this field defaults to external_model.name, with '.' and ':' replaced with '-', and if
3205
3207
  not specified for other entities, it defaults to entity_name-entity_version."""
3206
3208
 
3207
- workload_size: Optional[ServedModelInputWorkloadSize] = None
3209
+ workload_size: Optional[str] = None
3208
3210
  """The workload size of the served entity. The workload size corresponds to a range of provisioned
3209
3211
  concurrency that the compute autoscales between. A single unit of provisioned concurrency can
3210
3212
  process one request at a time. Valid workload sizes are "Small" (4 - 4 provisioned concurrency),
3211
- "Medium" (8 - 16 provisioned concurrency), and "Large" (16 - 64 provisioned concurrency). If
3213
+ "Medium" (8 - 16 provisioned concurrency), and "Large" (16 - 64 provisioned concurrency).
3214
+ Additional custom workload sizes can also be used when available in the workspace. If
3212
3215
  scale-to-zero is enabled, the lower bound of the provisioned concurrency for each workload size
3213
3216
  is 0."""
3214
3217
 
@@ -3240,7 +3243,7 @@ class ServedModelInput:
3240
3243
  if self.scale_to_zero_enabled is not None:
3241
3244
  body["scale_to_zero_enabled"] = self.scale_to_zero_enabled
3242
3245
  if self.workload_size is not None:
3243
- body["workload_size"] = self.workload_size.value
3246
+ body["workload_size"] = self.workload_size
3244
3247
  if self.workload_type is not None:
3245
3248
  body["workload_type"] = self.workload_type.value
3246
3249
  return body
@@ -3282,18 +3285,11 @@ class ServedModelInput:
3282
3285
  model_version=d.get("model_version", None),
3283
3286
  name=d.get("name", None),
3284
3287
  scale_to_zero_enabled=d.get("scale_to_zero_enabled", None),
3285
- workload_size=_enum(d, "workload_size", ServedModelInputWorkloadSize),
3288
+ workload_size=d.get("workload_size", None),
3286
3289
  workload_type=_enum(d, "workload_type", ServedModelInputWorkloadType),
3287
3290
  )
3288
3291
 
3289
3292
 
3290
- class ServedModelInputWorkloadSize(Enum):
3291
-
3292
- LARGE = "Large"
3293
- MEDIUM = "Medium"
3294
- SMALL = "Small"
3295
-
3296
-
3297
3293
  class ServedModelInputWorkloadType(Enum):
3298
3294
  """Please keep this in sync with with workload types in InferenceEndpointEntities.scala"""
3299
3295
 
@@ -3338,7 +3334,8 @@ class ServedModelOutput:
3338
3334
  """The workload size of the served entity. The workload size corresponds to a range of provisioned
3339
3335
  concurrency that the compute autoscales between. A single unit of provisioned concurrency can
3340
3336
  process one request at a time. Valid workload sizes are "Small" (4 - 4 provisioned concurrency),
3341
- "Medium" (8 - 16 provisioned concurrency), and "Large" (16 - 64 provisioned concurrency). If
3337
+ "Medium" (8 - 16 provisioned concurrency), and "Large" (16 - 64 provisioned concurrency).
3338
+ Additional custom workload sizes can also be used when available in the workspace. If
3342
3339
  scale-to-zero is enabled, the lower bound of the provisioned concurrency for each workload size
3343
3340
  is 0."""
3344
3341