@aws-sdk/client-sagemaker 3.1069.0 → 3.1070.0

package/dist-types/models/models_1.d.ts

@@ -89,6 +89,33 @@ export interface ConditionStepMetadata {
      */
     Outcome?: ConditionOutcome | undefined;
 }
+/**
+ * <p>Specifies a metrics endpoint for a container, including the path where the container exposes Prometheus-formatted metrics and the frequency at which to publish them to Amazon CloudWatch.</p>
+ * @public
+ */
+export interface MetricsEndpoint {
+    /**
+     * <p>The path to the metrics endpoint exposed by the container. For example, <code>/metrics</code> or <code>/server/metrics</code>. The path must start with <code>/</code> and can contain alphanumeric characters, forward slashes, underscores, hyphens, and periods. Maximum length is 256 characters. If not specified, defaults to <code>/metrics</code>.</p>
+     * @public
+     */
+    MetricsEndpointPath: string | undefined;
+    /**
+     * <p>The interval, in seconds, at which container metrics scraped from the endpoint are published to Amazon CloudWatch. Valid values: <code>10</code>, <code>30</code>, <code>60</code>, <code>120</code>, <code>180</code>, <code>240</code>, <code>300</code>. Defaults to <code>60</code>.</p>
+     * @public
+     */
+    MetricPublishFrequencyInSeconds?: MetricPublishFrequencyInSeconds | undefined;
+}
+/**
+ * <p>The configuration for container-level metrics scraping. Use this configuration to specify a custom metrics endpoint path and publishing frequency for container metrics. When <code>EnableDetailedObservability</code> is set to <code>True</code> in <code>MetricsConfig</code>, metrics are scraped from the container's Prometheus endpoint. If this configuration is not provided, the default path <code>/metrics</code> on port <code>8080</code> is used with a default publishing frequency of <code>60</code> seconds. For first-party and Deep Learning Containers (DLC), the endpoint path is determined automatically and this configuration is optional.</p>
+ * @public
+ */
+export interface ContainerMetricsConfig {
+    /**
+     * <p>A list of metrics endpoints to scrape from the container. Each endpoint specifies the path where the container exposes Prometheus-formatted metrics and the frequency at which to publish them. You can specify a maximum of 1 endpoint.</p>
+     * @public
+     */
+    MetricsEndpoints?: MetricsEndpoint[] | undefined;
+}
 /**
  * <p>Specifies an authentication configuration for the private docker registry where your model image is hosted. Specify a value for this property only if you specified <code>Vpc</code> as the value for the <code>RepositoryAccessMode</code> field of the <code>ImageConfig</code> object that you passed to a call to <code>CreateModel</code> and the private Docker registry where the model image is hosted requires authentication.</p>
  * @public
@@ -187,6 +214,11 @@ export interface ContainerDefinition {
      * @public
      */
     MultiModelConfig?: MultiModelConfig | undefined;
+    /**
+     * <p>The configuration for container metrics scraping. Specifies the metrics endpoint path and publishing frequency. If not specified when <code>EnableDetailedObservability</code> is <code>True</code>, the default path <code>/metrics</code> on port <code>8080</code> is used. For first-party and Deep Learning Containers (DLC), the endpoint path is determined automatically and this configuration is optional.</p>
+     * @public
+     */
+    ContainerMetricsConfig?: ContainerMetricsConfig | undefined;
 }
 /**
  * <p>A structure describing the source of a context.</p>
@@ -2913,7 +2945,12 @@ export interface MetricsConfig {
      */
     EnableEnhancedMetrics?: boolean | undefined;
     /**
-     * <p>The interval, in seconds, at which metrics are published to Amazon CloudWatch. Defaults to <code>60</code>. Valid values: <code>10</code>, <code>30</code>, <code>60</code>, <code>120</code>, <code>180</code>, <code>240</code>, <code>300</code>. When <code>EnableEnhancedMetrics</code> is set to <code>False</code>, this interval applies to utilization metrics only; invocation metrics continue to be published at the default 60-second interval. When <code>EnableEnhancedMetrics</code> is set to <code>True</code>, this interval applies to both utilization and invocation metrics.</p>
+     * <p>Indicates whether detailed observability is enabled for the endpoint. When set to <code>True</code>, the following metrics are published at the configured frequency:</p> <ul> <li> <p>Container-level inference metrics scraped from the container's Prometheus endpoint (such as request latency, error counts, and throughput). Available metrics vary by framework.</p> </li> <li> <p>Per-GPU metrics (utilization, memory, and temperature) attributed to individual inference components.</p> </li> <li> <p>Per-instance host metrics (CPU, memory, and disk utilization).</p> </li> <li> <p>Inference component placement metrics (copy count per Availability Zone).</p> </li> </ul> <p>For first-party and Deep Learning Containers (DLC), the Prometheus endpoint path is determined automatically. For Bring-Your-Own-Container (BYOC) cases, you can optionally set <code>ContainerMetricsConfig</code> to specify a custom endpoint path. If not specified, the default path <code>/metrics</code> on port <code>8080</code> is used.</p> <p>When set to <code>False</code>, these additional metrics are not published. Standard invocation and utilization metrics controlled by <code>EnableEnhancedMetrics</code> are unaffected.</p> <p>The default value for new endpoint configurations is <code>True</code>. For existing endpoint configurations created before this feature, the value is <code>False</code> unless explicitly set.</p>
+     * @public
+     */
+    EnableDetailedObservability?: boolean | undefined;
+    /**
+     * <p>The interval, in seconds, at which metrics are published to Amazon CloudWatch. Defaults to <code>60</code>. Valid values: <code>10</code>, <code>30</code>, <code>60</code>, <code>120</code>, <code>180</code>, <code>240</code>, <code>300</code>.</p> <p>When <code>EnableEnhancedMetrics</code> is set to <code>False</code>, this interval applies to utilization metrics only. Invocation metrics continue to be published at the default 60-second interval. When <code>EnableEnhancedMetrics</code> is set to <code>True</code>, this interval applies to both utilization and invocation metrics.</p> <p>When <code>EnableDetailedObservability</code> is set to <code>True</code>, this interval applies to per-GPU metrics, per-instance host metrics, container metrics, and fleet-level inference component lifecycle and placement metrics.</p>
      * @public
      */
     MetricPublishFrequencyInSeconds?: MetricPublishFrequencyInSeconds | undefined;
@@ -4485,6 +4522,11 @@ export interface InferenceComponentContainerSpecification {
      * @public
      */
     Environment?: Record<string, string> | undefined;
+    /**
+     * <p>The configuration for container metrics scraping. Specifies the metrics endpoint path and publishing frequency for the inference component's container. If not specified when <code>EnableDetailedObservability</code> is <code>True</code>, the default path <code>/metrics</code> on port <code>8080</code> is used. For first-party and Deep Learning Containers (DLC), the endpoint path is determined automatically and this configuration is optional.</p>
+     * @public
+     */
+    ContainerMetricsConfig?: ContainerMetricsConfig | undefined;
 }
 /**
  * <p>Settings that affect how the inference component caches data.</p>
@@ -7148,35 +7190,3 @@ export interface OptimizationJobModelSource {
      */
     SageMakerModel?: OptimizationSageMakerModel | undefined;
 }
-/**
- * <p>Settings for the model compilation technique that's applied by a model optimization job.</p>
- * @public
- */
-export interface ModelCompilationConfig {
-    /**
-     * <p>The URI of an LMI DLC in Amazon ECR. SageMaker uses this image to run the optimization.</p>
-     * @public
-     */
-    Image?: string | undefined;
-    /**
-     * <p>Environment variables that override the default ones in the model container.</p>
-     * @public
-     */
-    OverrideEnvironment?: Record<string, string> | undefined;
-}
-/**
- * <p>Settings for the model quantization technique that's applied by a model optimization job.</p>
- * @public
- */
-export interface ModelQuantizationConfig {
-    /**
-     * <p>The URI of an LMI DLC in Amazon ECR. SageMaker uses this image to run the optimization.</p>
-     * @public
-     */
-    Image?: string | undefined;
-    /**
-     * <p>Environment variables that override the default ones in the model container.</p>
-     * @public
-     */
-    OverrideEnvironment?: Record<string, string> | undefined;
-}

package/dist-types/models/models_2.d.ts

@@ -1,6 +1,38 @@
-import type { ActionStatus, ActivationState, AIBenchmarkJobStatus, AIRecommendationJobStatus, AlgorithmStatus, AppNetworkAccessType, AppSecurityGroupManagement, AppStatus, AppType, AuthMode, AutoMLJobSecondaryStatus, AutoMLJobStatus, AutoMLProblemTypeConfigName, BatchStrategy, CapacityReservationPreference, CaptureStatus, ClusterNodeProvisioningMode, ClusterNodeRecovery, ClusterStatus, CompilationJobStatus, CustomizationTechnique, DataDistributionType, DeepHealthCheckType, DomainStatus, EdgePackagingJobStatus, EdgePresetDeploymentStatus, EdgePresetDeploymentType, EnabledOrDisabled, EndpointStatus, EvaluationType, FeatureGroupStatus, FeatureStatus, FeatureType, FlowDefinitionStatus, HomeEfsFileSystemCreation, HubContentStatus, HubContentSupportStatus, HubContentType, HubStatus, HumanTaskUiStatus, HyperParameterTuningJobObjectiveType, HyperParameterTuningJobStatus, ImageStatus, ImageVersionStatus, InputMode, JobCategory, JobType, JoinSource, LastUpdateStatusValue, ModelSpeculativeDecodingS3DataType, ModelSpeculativeDecodingTechnique, ObjectiveStatus, OfflineStoreStatusValue, OptimizationJobDeploymentInstanceType, PartnerAppAuthType, PartnerAppType, Peft, ProblemType, ProcessingInstanceType, ProcessingS3CompressionType, ProcessingS3DataDistributionType, ProcessingS3DataType, ProcessingS3InputMode, ProcessingS3UploadMode, Processor, ProductionVariantAcceleratorType, ProductionVariantInstanceType, RecommendationStatus, RedshiftResultCompressionType, RedshiftResultFormat, RetentionType, RuleEvaluationStatus, SchedulerConfigComponent, SchedulerResourceStatus, ServerlessJobType, SharingType, StageStatus, Statistic, StudioLifecycleConfigAppType, TagPropagation, ThroughputMode, TrainingJobStatus, TrialComponentPrimaryStatus, VariantStatus, VendorGuidance, WorkforceIpAddressType } from "./enums";
+import type { ActionStatus, ActivationState, AIBenchmarkJobStatus, AIRecommendationJobStatus, AlgorithmStatus, AppNetworkAccessType, AppSecurityGroupManagement, AppStatus, AppType, AuthMode, AutoMLJobSecondaryStatus, AutoMLJobStatus, AutoMLProblemTypeConfigName, BatchStrategy, CapacityReservationPreference, CaptureStatus, ClusterNodeProvisioningMode, ClusterNodeRecovery, ClusterStatus, CompilationJobStatus, CustomizationTechnique, DataDistributionType, DeepHealthCheckType, DomainStatus, EdgePackagingJobStatus, EdgePresetDeploymentStatus, EdgePresetDeploymentType, EnabledOrDisabled, EndpointStatus, EvaluationType, FeatureGroupStatus, FeatureStatus, FeatureType, FlowDefinitionStatus, HomeEfsFileSystemCreation, HubContentStatus, HubContentSupportStatus, HubContentType, HubStatus, HumanTaskUiStatus, HyperParameterTuningJobObjectiveType, HyperParameterTuningJobStatus, ImageStatus, InputMode, JobCategory, JoinSource, LastUpdateStatusValue, ModelSpeculativeDecodingS3DataType, ModelSpeculativeDecodingTechnique, ObjectiveStatus, OfflineStoreStatusValue, OptimizationJobDeploymentInstanceType, PartnerAppAuthType, PartnerAppType, Peft, ProblemType, ProcessingInstanceType, ProcessingS3CompressionType, ProcessingS3DataDistributionType, ProcessingS3DataType, ProcessingS3InputMode, ProcessingS3UploadMode, ProductionVariantAcceleratorType, ProductionVariantInstanceType, RecommendationStatus, RedshiftResultCompressionType, RedshiftResultFormat, RetentionType, RuleEvaluationStatus, SchedulerConfigComponent, SchedulerResourceStatus, ServerlessJobType, SharingType, StageStatus, Statistic, StudioLifecycleConfigAppType, TagPropagation, ThroughputMode, TrainingJobStatus, TrialComponentPrimaryStatus, VariantStatus, WorkforceIpAddressType } from "./enums";
 import type { ActionSource, AIBenchmarkNetworkConfig, AIBenchmarkOutputResult, AIBenchmarkTarget, AIDatasetConfig, AIModelSource, AIRecommendation, AIRecommendationComputeSpec, AIRecommendationInferenceSpecification, AIRecommendationOutputResult, AIRecommendationPerformanceTarget, AIWorkloadConfigs, AlgorithmSpecification, AlgorithmStatusDetails, AlgorithmValidationSpecification, AppSpecification, ArtifactSource, AsyncInferenceConfig, AthenaDatasetDefinition, AutoMLCandidate, AutoMLChannel, AutoMLComputeConfig, AutoMLDataSplitConfig, AutoMLJobArtifacts, AutoMLJobChannel, AutoMLJobCompletionCriteria, AutoMLJobConfig, AutoMLJobObjective, AutoMLOutputDataConfig, AutoMLPartialFailureReason, AutoMLProblemTypeConfig, AutoMLResolvedAttributes, AutoMLSecurityConfig, Autotune, BatchDataCaptureConfig, CfnCreateTemplateProvider, Channel, CheckpointConfig, ClusterAutoScalingConfigOutput, ClusterEventDetail, ClusterInstanceGroupDetails, ClusterNodeDetails, ClusterOrchestrator, ClusterRestrictedInstanceGroupDetails, ClusterRestrictedInstanceGroupsConfigOutput, ClusterTieredStorageConfig, CodeEditorAppImageConfig, CodeRepository, CognitoConfig, CognitoMemberDefinition, CollectionConfiguration, ComputeQuotaConfig, GitConfig, InferenceSpecification, JupyterLabAppImageConfig, KernelGatewayImageConfig, OutputDataConfig, ResourceConfig, ResourceSpec, StoppingCondition, Tag, TransformInput, TransformOutput, TransformResources, UserContext, VpcConfig } from "./models_0";
-import type { ComputeQuotaTarget, ContextSource, DataCaptureConfig, DataQualityAppSpecification, DataQualityBaselineConfig, DataQualityJobInput, DefaultSpaceSettings, DeploymentConfig, DeviceSelectionConfig, DomainSettings, EdgeDeploymentConfig, EdgeDeploymentModelConfig, EdgeOutputConfig, ExplainerConfig, FeatureDefinition, FlowDefinitionOutputConfig, HubS3StorageConfig, HumanLoopActivationConfig, HumanLoopConfig, HumanLoopRequestSource, HyperParameterTrainingJobDefinition, HyperParameterTuningJobConfig, HyperParameterTuningJobWarmStartConfig, InputConfig, JupyterServerAppSettings, KernelGatewayAppSettings, MetadataProperties, MetricsConfig, ModelCompilationConfig, ModelDeployConfig, ModelQuantizationConfig, MonitoringNetworkConfig, MonitoringOutputConfig, MonitoringResources, MonitoringStoppingCondition, NeoVpcConfig, NetworkConfig, OfflineStoreConfig, OnlineStoreConfig, OptimizationJobModelSource, OptimizationSageMakerModel, OutputConfig, ProductionVariant, ProductionVariantManagedInstanceScaling, ProductionVariantRoutingConfig, ProductionVariantServerlessConfig, RetryStrategy, SchedulerConfig, TrainingSpecification, UserSettings } from "./models_1";
+import type { ComputeQuotaTarget, ContextSource, DataCaptureConfig, DataQualityAppSpecification, DataQualityBaselineConfig, DataQualityJobInput, DefaultSpaceSettings, DeploymentConfig, DeviceSelectionConfig, DomainSettings, EdgeDeploymentConfig, EdgeDeploymentModelConfig, EdgeOutputConfig, ExplainerConfig, FeatureDefinition, FlowDefinitionOutputConfig, HubS3StorageConfig, HumanLoopActivationConfig, HumanLoopConfig, HumanLoopRequestSource, HyperParameterTrainingJobDefinition, HyperParameterTuningJobConfig, HyperParameterTuningJobWarmStartConfig, InputConfig, JupyterServerAppSettings, KernelGatewayAppSettings, MetadataProperties, MetricsConfig, ModelDeployConfig, MonitoringNetworkConfig, MonitoringOutputConfig, MonitoringResources, MonitoringStoppingCondition, NeoVpcConfig, NetworkConfig, OfflineStoreConfig, OnlineStoreConfig, OptimizationJobModelSource, OptimizationSageMakerModel, OutputConfig, ProductionVariant, ProductionVariantManagedInstanceScaling, ProductionVariantRoutingConfig, ProductionVariantServerlessConfig, RetryStrategy, SchedulerConfig, TrainingSpecification, UserSettings } from "./models_1";
+/**
+ * <p>Settings for the model compilation technique that's applied by a model optimization job.</p>
+ * @public
+ */
+export interface ModelCompilationConfig {
+    /**
+     * <p>The URI of an LMI DLC in Amazon ECR. SageMaker uses this image to run the optimization.</p>
+     * @public
+     */
+    Image?: string | undefined;
+    /**
+     * <p>Environment variables that override the default ones in the model container.</p>
+     * @public
+     */
+    OverrideEnvironment?: Record<string, string> | undefined;
+}
+/**
+ * <p>Settings for the model quantization technique that's applied by a model optimization job.</p>
+ * @public
+ */
+export interface ModelQuantizationConfig {
+    /**
+     * <p>The URI of an LMI DLC in Amazon ECR. SageMaker uses this image to run the optimization.</p>
+     * @public
+     */
+    Image?: string | undefined;
+    /**
+     * <p>Environment variables that override the default ones in the model container.</p>
+     * @public
+     */
+    OverrideEnvironment?: Record<string, string> | undefined;
+}
 /**
  * <p>Settings for the model sharding technique that's applied by a model optimization job.</p>
  * @public
@@ -7100,108 +7132,3 @@ export interface DescribeImageResponse {
      */
     RoleArn?: string | undefined;
 }
-/**
- * @public
- */
-export interface DescribeImageVersionRequest {
-    /**
-     * <p>The name of the image.</p>
-     * @public
-     */
-    ImageName: string | undefined;
-    /**
-     * <p>The version of the image. If not specified, the latest version is described.</p>
-     * @public
-     */
-    Version?: number | undefined;
-    /**
-     * <p>The alias of the image version.</p>
-     * @public
-     */
-    Alias?: string | undefined;
-}
-/**
- * @public
- */
-export interface DescribeImageVersionResponse {
-    /**
-     * <p>The registry path of the container image on which this image version is based.</p>
-     * @public
-     */
-    BaseImage?: string | undefined;
-    /**
-     * <p>The registry path of the container image that contains this image version.</p>
-     * @public
-     */
-    ContainerImage?: string | undefined;
-    /**
-     * <p>When the version was created.</p>
-     * @public
-     */
-    CreationTime?: Date | undefined;
-    /**
-     * <p>When a create or delete operation fails, the reason for the failure.</p>
-     * @public
-     */
-    FailureReason?: string | undefined;
-    /**
-     * <p>The ARN of the image the version is based on.</p>
-     * @public
-     */
-    ImageArn?: string | undefined;
-    /**
-     * <p>The ARN of the version.</p>
-     * @public
-     */
-    ImageVersionArn?: string | undefined;
-    /**
-     * <p>The status of the version.</p>
-     * @public
-     */
-    ImageVersionStatus?: ImageVersionStatus | undefined;
-    /**
-     * <p>When the version was last modified.</p>
-     * @public
-     */
-    LastModifiedTime?: Date | undefined;
-    /**
-     * <p>The version number.</p>
-     * @public
-     */
-    Version?: number | undefined;
-    /**
-     * <p>The stability of the image version specified by the maintainer.</p> <ul> <li> <p> <code>NOT_PROVIDED</code>: The maintainers did not provide a status for image version stability.</p> </li> <li> <p> <code>STABLE</code>: The image version is stable.</p> </li> <li> <p> <code>TO_BE_ARCHIVED</code>: The image version is set to be archived. Custom image versions that are set to be archived are automatically archived after three months.</p> </li> <li> <p> <code>ARCHIVED</code>: The image version is archived. Archived image versions are not searchable and are no longer actively supported. </p> </li> </ul>
-     * @public
-     */
-    VendorGuidance?: VendorGuidance | undefined;
-    /**
-     * <p>Indicates SageMaker AI job type compatibility.</p> <ul> <li> <p> <code>TRAINING</code>: The image version is compatible with SageMaker AI training jobs.</p> </li> <li> <p> <code>INFERENCE</code>: The image version is compatible with SageMaker AI inference jobs.</p> </li> <li> <p> <code>NOTEBOOK_KERNEL</code>: The image version is compatible with SageMaker AI notebook kernels.</p> </li> </ul>
-     * @public
-     */
-    JobType?: JobType | undefined;
-    /**
-     * <p>The machine learning framework vended in the image version.</p>
-     * @public
-     */
-    MLFramework?: string | undefined;
-    /**
-     * <p>The supported programming language and its version.</p>
-     * @public
-     */
-    ProgrammingLang?: string | undefined;
-    /**
-     * <p>Indicates CPU or GPU compatibility.</p> <ul> <li> <p> <code>CPU</code>: The image version is compatible with CPU.</p> </li> <li> <p> <code>GPU</code>: The image version is compatible with GPU.</p> </li> </ul>
-     * @public
-     */
-    Processor?: Processor | undefined;
-    /**
-     * <p>Indicates Horovod compatibility.</p>
-     * @public
-     */
-    Horovod?: boolean | undefined;
-    /**
-     * <p>The maintainer description of the image version.</p>
-     * @public
-     */
-    ReleaseNotes?: string | undefined;
-}

package/dist-types/models/models_3.d.ts

@@ -1,7 +1,112 @@
-import type { _InstanceType, AccountDefaultStatus, AIBenchmarkJobStatus, AIRecommendationJobStatus, AlgorithmSortBy, AppImageConfigSortKey, AppSortKey, AssociationEdgeType, AutoMLJobStatus, AutoMLSortBy, AutoMLSortOrder, BatchStrategy, CandidateSortBy, CandidateStatus, ClusterEventResourceType, ClusterSortBy, CodeRepositorySortBy, CodeRepositorySortOrder, CompilationJobStatus, DetailedModelPackageStatus, DeviceDeploymentStatus, DirectInternetAccess, DomainStatus, EdgePackagingJobStatus, EndpointConfigSortKey, EndpointSortKey, EndpointStatus, EventSortBy, ExecutionRoleIdentityConfig, ExecutionStatus, FeatureGroupSortBy, FeatureGroupSortOrder, FeatureGroupStatus, FeatureType, FlowDefinitionStatus, HubContentSortBy, HubContentStatus, HubContentSupportStatus, HubContentType, HubSortBy, HubStatus, HyperParameterTuningJobSortByOptions, HyperParameterTuningJobStatus, HyperParameterTuningJobStrategyType, ImageSortBy, ImageSortOrder, ImageStatus, ImageVersionSortBy, ImageVersionSortOrder, ImageVersionStatus, IncludedData, InferenceComponentCapacitySizeType, InferenceComponentSortKey, InferenceComponentStatus, InferenceExperimentStatus, InferenceExperimentType, IPAddressType, IsTrackingServerActive, JobCategory, JobSecondaryStatus, JobStatus, LabelingJobStatus, ListAIBenchmarkJobsSortBy, ListAIRecommendationJobsSortBy, ListAIWorkloadConfigsSortBy, ListCompilationJobsSortBy, ListDeviceFleetsSortBy, ListEdgeDeploymentPlansSortBy, ListEdgePackagingJobsSortBy, ListInferenceRecommendationsJobsSortBy, MaintenanceStatus, ManagedStorageType, MlflowAppStatus, ModelApprovalStatus, ModelCardExportJobStatus, ModelCardProcessingStatus, ModelCardStatus, ModelPackageGroupStatus, ModelPackageRegistrationType, ModelPackageStatus, ModelRegistrationMode, ModelVariantStatus, MonitoringJobDefinitionSortKey, MonitoringType, NotebookInstanceAcceleratorType, NotebookInstanceStatus, OfflineStoreStatusValue, Operator, OptimizationJobDeploymentInstanceType, OptimizationJobStatus, OrderKey, PartnerAppAuthType, PartnerAppStatus, PartnerAppType, PipelineExecutionStatus, PipelineStatus, ProcessingJobStatus, ProductionVariantInstanceType, ProfilingStatus, ProjectStatus, RecommendationJobStatus, RecommendationJobType, RecommendationStepType, ReservedCapacityInstanceType, ReservedCapacityStatus, ReservedCapacityType, ResourceType, RootAccess, RuleEvaluationStatus, SageMakerResourceName, SagemakerServicecatalogStatus, SchedulerResourceStatus, ScheduleStatus, SecondaryStatus, SkipModelValidation, SortActionsBy, SortArtifactsBy, SortAssociationsBy, SortClusterSchedulerConfigBy, SortContextsBy, SortExperimentsBy, SortInferenceExperimentsBy, SortOrder, SortQuotaBy, SpaceStatus, StudioLifecycleConfigAppType, TrackingServerMaintenanceStatus, TrackingServerSize, TrackingServerStatus, TrainingJobStatus, TrainingPlanStatus, TransformJobStatus, UserProfileStatus, VolumeAttachmentStatus, WarmPoolResourceStatus, WorkforceIpAddressType, WorkforceStatus } from "./enums";
+import type { _InstanceType, AccountDefaultStatus, AIBenchmarkJobStatus, AIRecommendationJobStatus, AlgorithmSortBy, AppImageConfigSortKey, AppSortKey, AssociationEdgeType, AutoMLJobStatus, AutoMLSortBy, AutoMLSortOrder, BatchStrategy, CandidateSortBy, CandidateStatus, ClusterEventResourceType, ClusterSortBy, CodeRepositorySortBy, CodeRepositorySortOrder, CompilationJobStatus, DetailedModelPackageStatus, DeviceDeploymentStatus, DirectInternetAccess, DomainStatus, EdgePackagingJobStatus, EndpointConfigSortKey, EndpointSortKey, EndpointStatus, EventSortBy, ExecutionRoleIdentityConfig, ExecutionStatus, FeatureGroupSortBy, FeatureGroupSortOrder, FeatureGroupStatus, FeatureType, FlowDefinitionStatus, HubContentSortBy, HubContentStatus, HubContentSupportStatus, HubContentType, HubSortBy, HubStatus, HyperParameterTuningJobSortByOptions, HyperParameterTuningJobStatus, HyperParameterTuningJobStrategyType, ImageSortBy, ImageSortOrder, ImageStatus, ImageVersionSortBy, ImageVersionSortOrder, ImageVersionStatus, IncludedData, InferenceComponentCapacitySizeType, InferenceComponentSortKey, InferenceComponentStatus, InferenceExperimentStatus, InferenceExperimentType, IPAddressType, IsTrackingServerActive, JobCategory, JobSecondaryStatus, JobStatus, JobType, LabelingJobStatus, ListAIBenchmarkJobsSortBy, ListAIRecommendationJobsSortBy, ListAIWorkloadConfigsSortBy, ListCompilationJobsSortBy, ListDeviceFleetsSortBy, ListEdgeDeploymentPlansSortBy, ListEdgePackagingJobsSortBy, MaintenanceStatus, ManagedStorageType, MlflowAppStatus, ModelApprovalStatus, ModelCardExportJobStatus, ModelCardProcessingStatus, ModelCardStatus, ModelPackageGroupStatus, ModelPackageRegistrationType, ModelPackageStatus, ModelRegistrationMode, ModelVariantStatus, MonitoringJobDefinitionSortKey, MonitoringType, NotebookInstanceAcceleratorType, NotebookInstanceStatus, OfflineStoreStatusValue, Operator, OptimizationJobDeploymentInstanceType, OptimizationJobStatus, OrderKey, PartnerAppAuthType, PartnerAppStatus, PartnerAppType, PipelineExecutionStatus, PipelineStatus, ProcessingJobStatus, Processor, ProductionVariantInstanceType, ProfilingStatus, ProjectStatus, RecommendationJobStatus, RecommendationJobType, RecommendationStepType, ReservedCapacityInstanceType, ReservedCapacityStatus, ReservedCapacityType, ResourceType, RootAccess, RuleEvaluationStatus, SageMakerResourceName, SagemakerServicecatalogStatus, SchedulerResourceStatus, ScheduleStatus, SecondaryStatus, SkipModelValidation, SortActionsBy, SortArtifactsBy, SortAssociationsBy, SortClusterSchedulerConfigBy, SortContextsBy, SortExperimentsBy, SortInferenceExperimentsBy, SortOrder, SortQuotaBy, SpaceStatus, StudioLifecycleConfigAppType, TrackingServerMaintenanceStatus, TrackingServerSize, TrackingServerStatus, TrainingJobStatus, TrainingPlanStatus, TransformJobStatus, UserProfileStatus, VendorGuidance, VolumeAttachmentStatus, WarmPoolResourceStatus, WorkforceIpAddressType, WorkforceStatus } from "./enums";
 import type { ActionSummary, AdditionalInferenceSpecificationDefinition, AgentVersion, AIBenchmarkJobSummary, AIRecommendationJobSummary, AIWorkloadConfigSummary, AlgorithmSpecification, AlgorithmSummary, AmazonQSettings, AppDetails, AppImageConfigDetails, AppSpecification, ArtifactSummary, AssociationInfo, AssociationSummary, AutoMLCandidate, AutoMLJobSummary, AutoRollbackConfig, AvailableUpgrade, BatchDataCaptureConfig, CfnTemplateProviderDetail, Channel, CheckpointConfig, ClusterEventSummary, ClusterNodeSummary, ClusterSchedulerConfigSummary, ClusterSummary, CodeRepositorySummary, CognitoConfig, CompilationJobSummary, InferenceSpecification, OutputDataConfig, OutputParameter, ResourceConfig, ResourceSpec, StoppingCondition, Tag, TransformInput, TransformOutput, TransformResources, UserContext, VpcConfig } from "./models_0";
-import type { ComputeQuotaSummary, ContainerDefinition, ContextSummary, DockerSettings, DriftCheckBaselines, EdgeOutputConfig, EndpointInfo, FeatureDefinition, HumanTaskConfig, HyperParameterTrainingJobDefinition, HyperParameterTuningJobConfig, HyperParameterTuningJobWarmStartConfig, InferenceComponentComputeResourceRequirements, InferenceComponentSchedulingConfig, InferenceComponentStartupParameters, InferenceExecutionConfig, InferenceExperimentDataStorageConfig, InferenceExperimentSchedule, InstanceMetadataServiceConfiguration, LabelingJobAlgorithmsConfig, LabelingJobInputConfig, LabelingJobOutputConfig, LabelingJobStoppingConditions, ManagedConfiguration, MetadataProperties, ModelBiasAppSpecification, ModelBiasBaselineConfig, ModelBiasJobInput, ModelCardExportOutputConfig, ModelCardSecurityConfig, ModelExplainabilityAppSpecification, ModelExplainabilityBaselineConfig, ModelExplainabilityJobInput, ModelInfrastructureConfig, ModelLifeCycle, ModelMetrics, ModelPackageModelCard, ModelPackageSecurityConfig, ModelPackageValidationSpecification, ModelQualityAppSpecification, ModelQualityBaselineConfig, ModelQualityJobInput, MonitoringNetworkConfig, MonitoringOutputConfig, MonitoringResources, MonitoringScheduleConfig, MonitoringStoppingCondition, NetworkConfig, NotebookInstanceLifecycleHook, OfflineStoreConfig, OnlineStoreConfig, OptimizationJobModelSource, ProductionVariantServerlessConfig, RecommendationJobInputConfig, RecommendationJobStoppingConditions, ResourceLimits, RetryStrategy, ShadowModeConfig, SourceAlgorithmSpecification, TrustedIdentityPropagationSettings, UnifiedStudioSettings, UserSettings } from "./models_1";
+import type { ComputeQuotaSummary, ContainerDefinition, ContainerMetricsConfig, ContextSummary, DockerSettings, DriftCheckBaselines, EdgeOutputConfig, EndpointInfo, FeatureDefinition, HumanTaskConfig, HyperParameterTrainingJobDefinition, HyperParameterTuningJobConfig, HyperParameterTuningJobWarmStartConfig, InferenceComponentComputeResourceRequirements, InferenceComponentSchedulingConfig, InferenceComponentStartupParameters, InferenceExecutionConfig, InferenceExperimentDataStorageConfig, InferenceExperimentSchedule, InstanceMetadataServiceConfiguration, LabelingJobAlgorithmsConfig, LabelingJobInputConfig, LabelingJobOutputConfig, LabelingJobStoppingConditions, ManagedConfiguration, MetadataProperties, ModelBiasAppSpecification, ModelBiasBaselineConfig, ModelBiasJobInput, ModelCardExportOutputConfig, ModelCardSecurityConfig, ModelExplainabilityAppSpecification, ModelExplainabilityBaselineConfig, ModelExplainabilityJobInput, ModelInfrastructureConfig, ModelLifeCycle, ModelMetrics, ModelPackageModelCard, ModelPackageSecurityConfig, ModelPackageValidationSpecification, ModelQualityAppSpecification, ModelQualityBaselineConfig, ModelQualityJobInput, MonitoringNetworkConfig, MonitoringOutputConfig, MonitoringResources, MonitoringScheduleConfig, MonitoringStoppingCondition, NetworkConfig, NotebookInstanceLifecycleHook, OfflineStoreConfig, OnlineStoreConfig, OptimizationJobModelSource, ProductionVariantServerlessConfig, RecommendationJobInputConfig, RecommendationJobStoppingConditions, ResourceLimits, RetryStrategy, ShadowModeConfig, SourceAlgorithmSpecification, TrustedIdentityPropagationSettings, UnifiedStudioSettings, UserSettings } from "./models_1";
 import type { CustomizedMetricSpecification, DataCaptureConfigSummary, DataProcessing, DebugHookConfig, DebugRuleConfiguration, DebugRuleEvaluationStatus, DeployedImage, DeploymentRecommendation, ExperimentConfig, ExperimentSource, FeatureParameter, HyperParameterTrainingJobSummary, HyperParameterTuningJobCompletionDetails, HyperParameterTuningJobConsumedResources, InfraCheckConfig, LastUpdateStatus, MemberDefinition, MlflowConfig, ModelArtifacts, ModelClientConfig, ModelPackageConfig, NotificationConfiguration, ObjectiveStatusCounters, OfflineStoreStatus, OptimizationConfig, OptimizationJobOutputConfig, OptimizationVpcConfig, OwnershipSettings, ParallelismConfiguration, PartnerAppConfig, PartnerAppMaintenanceConfig, ProcessingInput, ProcessingOutputConfig, ProcessingResources, ProcessingStoppingCondition, ProductionVariantSummary, ProfilerConfig, ProfilerRuleConfiguration, RemoteDebugConfig, ServerlessJobConfig, ServiceCatalogProvisioningDetails, SourceIpConfig, SpaceSettings, SpaceSharingSettings, TensorBoardOutputConfig, TrainingJobStatusCounters, TrialComponentArtifact, TrialComponentParameterValue, TrialComponentStatus, WorkerAccessConfiguration } from "./models_2";
+/**
+ * @public
+ */
+export interface DescribeImageVersionRequest {
+    /**
+     * <p>The name of the image.</p>
+     * @public
+     */
+    ImageName: string | undefined;
+    /**
+     * <p>The version of the image. If not specified, the latest version is described.</p>
+     * @public
+     */
+    Version?: number | undefined;
+    /**
+     * <p>The alias of the image version.</p>
+     * @public
+     */
+    Alias?: string | undefined;
+}
+/**
+ * @public
+ */
+export interface DescribeImageVersionResponse {
+    /**
+     * <p>The registry path of the container image on which this image version is based.</p>
+     * @public
+     */
+    BaseImage?: string | undefined;
+    /**
+     * <p>The registry path of the container image that contains this image version.</p>
+     * @public
+     */
+    ContainerImage?: string | undefined;
+    /**
+     * <p>When the version was created.</p>
+     * @public
+     */
+    CreationTime?: Date | undefined;
+    /**
+     * <p>When a create or delete operation fails, the reason for the failure.</p>
+     * @public
+     */
+    FailureReason?: string | undefined;
+    /**
+     * <p>The ARN of the image the version is based on.</p>
+     * @public
+     */
+    ImageArn?: string | undefined;
+    /**
+     * <p>The ARN of the version.</p>
+     * @public
+     */
+    ImageVersionArn?: string | undefined;
+    /**
+     * <p>The status of the version.</p>
+     * @public
+     */
+    ImageVersionStatus?: ImageVersionStatus | undefined;
+    /**
+     * <p>When the version was last modified.</p>
+     * @public
+     */
+    LastModifiedTime?: Date | undefined;
+    /**
+     * <p>The version number.</p>
+     * @public
+     */
+    Version?: number | undefined;
+    /**
+     * <p>The stability of the image version specified by the maintainer.</p> <ul> <li> <p> <code>NOT_PROVIDED</code>: The maintainers did not provide a status for image version stability.</p> </li> <li> <p> <code>STABLE</code>: The image version is stable.</p> </li> <li> <p> <code>TO_BE_ARCHIVED</code>: The image version is set to be archived. Custom image versions that are set to be archived are automatically archived after three months.</p> </li> <li> <p> <code>ARCHIVED</code>: The image version is archived. Archived image versions are not searchable and are no longer actively supported. </p> </li> </ul>
+     * @public
+     */
+    VendorGuidance?: VendorGuidance | undefined;
+    /**
+     * <p>Indicates SageMaker AI job type compatibility.</p> <ul> <li> <p> <code>TRAINING</code>: The image version is compatible with SageMaker AI training jobs.</p> </li> <li> <p> <code>INFERENCE</code>: The image version is compatible with SageMaker AI inference jobs.</p> </li> <li> <p> <code>NOTEBOOK_KERNEL</code>: The image version is compatible with SageMaker AI notebook kernels.</p> </li> </ul>
+     * @public
+     */
+    JobType?: JobType | undefined;
+    /**
+     * <p>The machine learning framework vended in the image version.</p>
+     * @public
+     */
+    MLFramework?: string | undefined;
+    /**
+     * <p>The supported programming language and its version.</p>
+     * @public
+     */
+    ProgrammingLang?: string | undefined;
+    /**
+     * <p>Indicates CPU or GPU compatibility.</p> <ul> <li> <p> <code>CPU</code>: The image version is compatible with CPU.</p> </li> <li> <p> <code>GPU</code>: The image version is compatible with GPU.</p> </li> </ul>
+     * @public
+     */
+    Processor?: Processor | undefined;
+    /**
+     * <p>Indicates Horovod compatibility.</p>
+     * @public
+     */
+    Horovod?: boolean | undefined;
+    /**
+     * <p>The maintainer description of the image version.</p>
+     * @public
+     */
+    ReleaseNotes?: string | undefined;
+}
 /**
  * @public
  */
@@ -127,6 +232,11 @@ export interface InferenceComponentContainerSpecificationSummary {
      * @public
      */
     Environment?: Record<string, string> | undefined;
+    /**
+     * <p>The container metrics scraping configuration for this inference component, including the metrics endpoint path and publishing frequency.</p>
+     * @public
+     */
+    ContainerMetricsConfig?: ContainerMetricsConfig | undefined;
 }
 /**
  * <p>Settings that affect how the inference component caches data.</p>
@@ -6912,8 +7022,7 @@ export interface InferenceRecommendationsJobStep {
     InferenceBenchmark?: RecommendationJobInferenceBenchmark | undefined;
 }
 /**
- * Search shape for Job. Mirrors DescribeJobResponse fields.
- * If you update DescribeJobResponse, update this structure as well.
+ * <p>The properties of a job returned by the <a href="https://docs.aws.amazon.com/sagemaker/latest/APIReference/API_Search.html">Search</a> API.</p>
  * @public
  */
 export interface Job {
@@ -9753,83 +9862,3 @@ export interface ListInferenceExperimentsResponse {
      */
     NextToken?: string | undefined;
 }
-/**
- * @public
- */
-export interface ListInferenceRecommendationsJobsRequest {
-    /**
-     * <p>A filter that returns only jobs created after the specified time (timestamp).</p>
-     * @public
-     */
-    CreationTimeAfter?: Date | undefined;
-    /**
-     * <p>A filter that returns only jobs created before the specified time (timestamp).</p>
-     * @public
-     */
-    CreationTimeBefore?: Date | undefined;
-    /**
-     * <p>A filter that returns only jobs that were last modified after the specified time (timestamp).</p>
-     * @public
-     */
-    LastModifiedTimeAfter?: Date | undefined;
-    /**
-     * <p>A filter that returns only jobs that were last modified before the specified time (timestamp).</p>
-     * @public
-     */
-    LastModifiedTimeBefore?: Date | undefined;
-    /**
-     * <p>A string in the job name. This filter returns only recommendations whose name contains the specified string.</p>
-     * @public
-     */
-    NameContains?: string | undefined;
-    /**
-     * <p>A filter that retrieves only inference recommendations jobs with a specific status.</p>
-     * @public
-     */
-    StatusEquals?: RecommendationJobStatus | undefined;
-    /**
-     * <p>The parameter by which to sort the results.</p>
-     * @public
-     */
-    SortBy?: ListInferenceRecommendationsJobsSortBy | undefined;
-    /**
-     * <p>The sort order for the results.</p>
-     * @public
-     */
-    SortOrder?: SortOrder | undefined;
-    /**
-     * <p>If the response to a previous <code>ListInferenceRecommendationsJobsRequest</code> request was truncated, the response includes a <code>NextToken</code>. To retrieve the next set of recommendations, use the token in the next request.</p>
-     * @public
-     */
-    NextToken?: string | undefined;
-    /**
-     * <p>The maximum number of recommendations to return in the response.</p>
-     * @public
-     */
-    MaxResults?: number | undefined;
-    /**
-     * <p>A filter that returns only jobs that were created for this model.</p>
-     * @public
-     */
-    ModelNameEquals?: string | undefined;
-    /**
-     * <p>A filter that returns only jobs that were created for this versioned model package.</p>
-     * @public
-     */
-    ModelPackageVersionArnEquals?: string | undefined;
-}
-/**
- * @public
- */
-export interface ListInferenceRecommendationsJobsResponse {
-    /**
-     * <p>The recommendations created from the Amazon SageMaker Inference Recommender job.</p>
-     * @public
-     */
-    InferenceRecommendationsJobs: InferenceRecommendationsJob[] | undefined;
-    /**
-     * <p>A token for getting the next set of recommendations, if there are any.</p>
-     * @public
-     */
-    NextToken?: string | undefined;
-}