NVIDIA Run:ai Policy API
Policies allow administrators to impose restrictions and set default values for researcher workloads. Restrictions and default values can be placed on CPUs, GPUs, and other resources or entities. For more information, see [Policies](https://run-ai-docs.nvidia.com/saas/platform-management/policies/#introduction).
NVIDIA Run:ai Policy API is one of 62 APIs that NVIDIA Run:ai publishes on the APIs.io network, described by a machine-readable OpenAPI specification.
Tagged areas include Policy. The published artifact set on APIs.io includes an OpenAPI specification, API documentation, and authentication docs.
This API exposes 25 operations across 7 paths, and defines 378 schemas. It is described by OpenAPI 3.0.3, at version latest.
Requests are made against a single base URL, https://app.run.ai.
25 operations
7 paths
378 schemas
6 DELETE7 GET6 PATCH6 PUT
Metadata
The identity and technical contract details declared by the specification.
Authentication & Security 1
NVIDIA Run:ai Policy API declares
1 security scheme
for authenticating requests.
It accepts HTTP bearer tokens (JWT) (bearerAuth).
By default, every request must be authenticated.
bearerAuth— Bearer authentication
Paths & Operations 25
Across 7 paths, the API surfaces 25 operations — 6 DELETE, 7 GET, 6 PATCH, 6 PUT. Each is listed below with its method, path, parameters, and response codes.
Policy 25
Policies allow administrators to impose restrictions and set default values for researcher workloads. Restrictions and default values can be placed on CPUs, GPUs, and other resour…
GET
/api/v2/policy
List policies
PATCH
/api/v2/policy/trainings
Update a training policy.
PUT
/api/v2/policy/trainings
Overwrite a training policy.
GET
/api/v2/policy/trainings
Get a training policy.
DELETE
/api/v2/policy/trainings
Delete a training policy.
PATCH
/api/v2/policy/workspaces
Update a workspace policy.
PUT
/api/v2/policy/workspaces
Overwrite a workspace policy.
GET
/api/v2/policy/workspaces
Get a workspace policy.
DELETE
/api/v2/policy/workspaces
Delete a workspace policy.
PATCH
/api/v2/policy/distributed
Update a distributed policy.
PUT
/api/v2/policy/distributed
Overwrite a distributed policy.
GET
/api/v2/policy/distributed
Get a distributed policy.
DELETE
/api/v2/policy/distributed
Delete a distributed policy.
PATCH
/api/v2/policy/inferences
Update an inference policy.
PUT
/api/v2/policy/inferences
Overwrite an inference policy.
GET
/api/v2/policy/inferences
Get an inference policy.
DELETE
/api/v2/policy/inferences
Delete an inference policy.
PATCH
/api/v2/policy/distributed-inferences
Update a distributed inference policy.
PUT
/api/v2/policy/distributed-inferences
Overwrite a distributed inference policy.
GET
/api/v2/policy/distributed-inferences
Get a distributed inference policy.
DELETE
/api/v2/policy/distributed-inferences
Delete a distributed inference policy.
PATCH
/api/v2/policy/nim-services
Update NVIDIA NIM service policy.
PUT
/api/v2/policy/nim-services
Overwrite NVIDIA NIM service policy.
GET
/api/v2/policy/nim-services
Get NVIDIA NIM service policy.
DELETE
/api/v2/policy/nim-services
Delete NVIDIA NIM service policy.
Schemas 378
The contract defines 378 schemas that model the data the API accepts and returns. The most detailed are ReplicaRulesV2 (32 properties), ReplicaDefaultsV2 (32 properties), DistributedInferenceLeaderWorkerDefaultsV2 (18 properties), DistributedInferenceLeaderWorkerRulesV2 (18 properties). Each schema is shown below with its type and property counts.
S3Common
object
3 properties
SecretRules
object
3 properties
PortServiceTypeOptions
object
1 property
AutoScalingMetricField
object
1 property
TrainingPolicyRulesV2
NodeAffinityRequired
object
If the affinity requirements specified by this field are not met at scheduling time, the pod will not be scheduled onto the node. If the affinity requirements…
1 property
DistributedInferenceLeaderWorkerRulesV2
object
18 properties
PortsRules
object
2 properties
Ports
array
Set of container ports that the workload exposes.
ProbesRules
object
1 property
UidGidSourceRules
object
HostPath
object
4 properties
ImagePullPolicyRules
object
ServingPortServiceType
string
The type of Kubernetes service to create for the inference deployment. Options include 'ClusterIP' (default), 'NodePort', 'LoadBalancer', and 'ExternalName'.
HostPathsDefaults
object
2 properties
RunAsGid
integer
The group id to run the entrypoint of the container which executes the workspace. Default to the value specified in the environment asset runAsGid field (optio…
Annotations
array
Set of annotations to populate into the container running the workload.
AutoScalingMetric
string
The metric to use for autoscaling. Mandatory if minReplicas < maxReplicas, except for the special case where minReplicas is set to 0 and maxReplicas is set to…
TolerationEffect
string
The taint effect to match. (mandatory)
ImagePullPolicy
string
Image pull policy. Defaults to Always if :latest tag is specified, otherwise it is IfNotPresent.
PolicyValidationStatus
object
1 property
DistributedInferenceRestartPolicyRules
object
ServingPortServiceTypeRules
object
ServingPortExposedProtocolOptions
object
1 property
PolicyListResponse
object
1 property
EnvironmentVariables
array
Set of environment variables to populate into the container running the workload.
CpuCoreRequest
number
CPU units to allocate for the created workload (0.5, 1, .etc). The workload will receive at least this amount of CPU. Note that the workload will not be schedu…
AutoScalingField
object
1 property
InitialReplicasField
object
1 property
ExtendedResourcesRules
object
2 properties
Parallelism
integer
Used with Hyperparameter Optimization. Specifies the maximum number of pods the workload should run at any given time (applicable to standard training only).
EmptyDirsDefaults
object
2 properties
NimServicePolicy
object
4 properties
DistributedInferenceImposedAssets
object
2 properties
Capability
string
NimServiceRules
DistributedInferenceReplicasField
object
1 property
StringRulesCommon
object
3 properties
PolicyListEntry
object
3 properties
DistributedPolicyDefaultsAndRulesV2
object
4 properties
WorkspacePolicyV2
object
4 properties
S3sDefaults
object
2 properties
DistributedFramework
string
The distributed training framework used in the workload.
NimServiceServingPortRules
object
A port for accessing the inference service
7 properties
ActivationReplicasField
object
1 property
ProbeHandlerScheme
string
Scheme to use for connecting to the host, defaults to HTTP.
DistributedInferenceStartupPolicyOptions
object
1 property
WorkingDir
string
Container's working directory. If not specified, the container runtime default will be used. This may be configured in the container image.
AutoScalingMetricRules
object
ServingPortExposedProtocol
string
The protocol to use for the exposed URL. If grpcPort is set, this defaults to grpc. Otherwise, it defaults to http.
DistributedInferenceStartupPolicyField
object
1 property
PvcInstance
object
PolicyMeta
DistributedPolicyChangeRequestV2
object
3 properties
AutoDeletionTimeAfterPreemption
integer
Specifies the duration after which a finished workload (completed or failed) will be automatically deleted. The default is 30 days. Log retention is managed se…
StringRules
object
EmptyDirItems
array
A list of emptyDir volumes to mount in the workload.
TrainingPolicyDefaultsV2
ExtendedResourceRules
object
Rules for the attributes of extendedResources
1 property
PvcsRules
object
2 properties
ScaleDownDelayField
object
1 property
DistributedInferenceServingPortContainerAndProtocol
object
2 properties
PolicyId
string
The unique id of the policy.
IntegerRulesOptional
object
6 properties
GitCommon
object
4 properties
HostPathRules
object
4 properties
TolerationsDefaults
object
2 properties
Port
object
A port for accessing the workload.
7 properties
AutoScalingMetricThreshold
integer
The threshold to use with the specified metric for autoscaling (mandatory).
ServingPortAccess
object
4 properties
BackoffLimit
integer
Specifies the number of retries before marking a workload as failed (not applicable to Inference workloads). The default value is 6.
PodAffinityTypeOptions
object
1 property
DistributedInferenceCommonDefaultsV2
object
RestartPolicyRule
object
HostIpc
boolean
Whether to enable host IPC. Defaults to false.
RelatedUrlsDefaults
object
2 properties
Tty
boolean
Whether this container should allocate a TTY for itself, also requires 'stdin' to be true.
Category
string
Specify the workload category assigned to the workload. Categories are used to classify and monitor different types of workloads within the NVIDIA Run:ai platf…
NumberRules
object
7 properties
S3sRules
object
2 properties
PortServiceTypeRules
object
ExposedUrl
object
A URL for accessing the workload.
9 properties
ServingPortServiceTypeOptions
object
1 property
NfssDefaults
object
2 properties
Error
object
3 properties
2 required
ExcludeField
object
1 property
TerminationGracePeriod
integer
Duration in seconds the pod needs to terminate gracefully upon probe failure. The grace period is the duration in seconds after the processes running in the po…
GitRules
object
4 properties
PvcClaimSize
string
Requested size for the PVC. Mandatory when existingPvc is false. Recommended sizes: TB/GB/MB/TIB/GIB/MIB
PvcAddedAttrValues
array
an optional array of key-values pairs that are written as annotations on the created PVC. the allowed attributes are determined according to the storage class…
DistributedInferenceLeaderDefaults
object
1 property
InferenceFields
object
PodAffinityRules
object
2 properties
Toleration
object
Toleration details.
7 properties
ClusterId
string
The id of the cluster.
S3AccessKey
object
3 properties
SecretFieldsNonUpdatable
object
1 property
NimServicePolicyDefaultsAndRules
object
3 properties
HostNetwork
boolean
Whether to enable host networking. Default to false.
DistributedInferenceCommonRulesV2
object
ServingPortExposeExternally
boolean
Indicates whether the inference serving endpoint should be accessible outside the cluster. If set to true, the endpoint will be exposed externally. To enable e…
GitsDefaults
object
2 properties
InferenceFieldsRules
object
2 properties
CleanPodPolicyRules
object
ServingPort
object
A port for accessing the inference service
WorkspacePolicyRulesV2
AutoScalingMetricNim
string
The metric to use for autoscaling (mandatory).
DistributedInferenceRestartPolicy
string
Determines the behavior when a pod fails. - RecreateGroupOnPodRestart: Restarts all pods in the group if any pod fails. - None: No automatic restart behavior i…
ServingPortContainerAndProtocol
object
2 properties
MpiLauncherCreationPolicyRules
object
RelatedUrls
array
Set of URLs that are related to the workload.
DataVolume
object
2 properties
ConfigMapItems
array
Set of config map volumes to use in the workload
CleanPodPolicy
string
The policy describes how to deal with pods when the job is finished.
PolicyCreationFields
object
5 properties
2 required
NimServicePolicyChangeRequest
object
3 properties
EnvironmentVariableUserCredential
object
Defines a reference to a user-created credential and a specific key within that credential whose value will populate the environment variable. User credentials…
2 properties
ServingPortGrpcPort
integer
The GRPC port that the container running the inference service exposes.
NimServicePolicyOverwriteRequest
object
2 properties
NimCacheRules
object
2 properties
AutoScalingCommonFields
object
NodePools
array
A prioritized list of node pools for the scheduler to run the workload on. The scheduler will always try to use the first node pool before moving to the next o…
ServingPortExposedUrl
string
The custom URL to use for the serving port. If empty (default), an autogenerated URL will be used.
PvcAddedAttrValue
object
2 properties
1 required
WorkspacePolicyDefaultsV2
SecretItems1
array
Set of secret volumes to use in the workload
MetricThresholdPercentageField
object
1 property
NfsItems
array
Set of nfs volumes to use in the workload.
ServingPortAccessAuthorizationTypeEnum
string
Specifies who can send inference requests to the serving endpoint: Possible values: - public: No authorization is required. (Default) - authenticatedUsers: Any…
GpuRequestRules
object
EnvironmentVariablePodFieldReference
object
Details of the field-reference and key use to populate the environment variable
1 property
MatchExpression
object
A selector that contains values, a key, and an operator that relates the key and values.
3 properties
2 required
GpuRequestType
string
Sets the unit type for GPU resources requests. Stated in terms of portion or memory. Sets the unit type for other GPU request fields. If gpuDevicesRequest 1, o…
EmptyDir
object
3 properties
DistributedInferenceServingPortProtocol
string
The protocol used to access the port.
PodAffinityType
string
The affinity type, required or preferred. (mandatory)
Command
string
A command to the server as the entry point of the container running the workload.
NimServiceMultiNode
object
Defines whether the NIM service runs as a multi-node deployment. If workers is set to 1 or more, the service runs in multi-node.
1 property
GitsRules
object
2 properties
DataVolumeInstance
object
Args
string
Arguments to the command that the container running the workload executes.
ServingConfiguration
object
The inference workload serving configuration.
2 properties
NodeType2
string
Nodes (machines), or a group of nodes on which the workload will run. To use this feature, your Administrator will need to label nodes. For more information, s…
ServingPortAuthorizationTypeOptions
object
1 property
ClaimInfoRules
object
3 properties
TolerationOperatorRules
object
InferencePolicyDefaultsAndRulesV2
object
4 properties
PreemptibilityRules
object
WorkspacePolicyDefaultsAndRulesV2
object
4 properties
ImagePullPolicyOptions
object
1 property
DistributedPolicyV2
object
4 properties
StringRulesOptions
object
1 property
ServingConfigurationField
object
1 property
RestartPolicy
string
Specify the restart policy of the workload pods. Default is empty, which is determine by the framework default
TolerationsRules
object
2 properties
DefaultMode
string
File permission mode in octal string format. This value must be a 4-digit octal number, representing the default file mode when mounting a Secret or ConfigMap…
DistributedInferenceLeaderRules
object
1 property
DistributedInferenceRulesV2
ExtendedResource
object
Quantity of an extended resource.
3 properties
InstancesRules
object
1 property
AnnotationsDefaults
object
1 property
DistributedInferenceWorkerRules
object
1 property
WorkspacePolicyChangeRequestV2
object
3 properties
CleanPodPolicyOptions
object
1 property
PolicyName
string
A chosen name for the policy.
SeccompProfileTypeOptions
object
1 property
NimServiceNgcAuthSecret
string
The name of a Kubernetes secret containing the NGC access credentials. The secret must contain a key named NGCAPIKEY with the API key as the value.
ScopeType
string
Probes
object
Probes are used to determine if the container is healthy and ready to accept traffic.
1 property
EmptyDirInstance
object
DistributedInferenceWorkersFieldRules
object
1 property
DistributedInferencePolicyOverwriteRequestV2
object
2 properties
SecretsDefaults
object
2 properties
PortRules
object
Rules for port attributes
6 properties
ClusterIdOptional
string
The id of the cluster.
RunAsUid
integer
The user id to run the entrypoint of the container which executes the workspace. Default to the value specified in the environment asset runAsUid field (option…
PvcFieldsUpdatable
object
1 property
TrainingPolicyV2
object
4 properties
PvcFieldsNonUpdatable
object
6 properties
Probe
object
6 properties
ImposedAssets
array
a list of datasource assets to import into the policy. these datasources will be included in any workload that will be created in the scope of the policy.
EnvironmentVariablesDefaults
object
1 property
Labels
array
Set of labels to populate into the container running the workload.
ProbeRules
object
6 properties
DistributedInferenceServingPortAuthorizationTypeOptions
object
1 property
DistributedInferenceDefaultsV2
SeccompProfileTypeRules
object
Stdin
boolean
Whether this container should allocate a buffer for stdin in the container runtime. If this is not set, reads from stdin in the container will always result in…
Capabilities
array
Add POSIX capabilities to running containers. Defaults to the default set of capabilities granted by the container runtime.
ImagePullSecretsDefaults
object
1 property
InferencesMaxReplicasField
object
1 property
EnvironmentVariable
object
Details of an environment variable which is populated into the container.
8 properties
ServingPortAuthorizationTypeRules
object
InferencePolicyChangeRequestV2
object
3 properties
Timestamp
string
The time at which an event occurred. e.g. the time an effective policy has been updated
ProjectIdOptional
integer
The id of the project.
ProbeHandler
object
The action taken to determine the health of the container. (mandatory)
1 property
TrainingPolicyChangeRequestV2
object
3 properties
GitPassword
object
3 properties
ImagePullSecretsRulesAttributes
object
Rules for the attributes of imagePullSecrets
2 properties
ArrayRules
object
4 properties
HostPathItems
array
Set of host paths to use in the workload.
ClaimInfo
object
Claim information for the newly created PVC. The information should not be provided when attempting to use existing PVC.
5 properties
DistributedInferenceServingPortAuthorizationTypeRules
object
DistributedInferenceServingPort
object
Defines the configuration for the inference serving endpoint. This determines how applications or services can send inference requests to the workload.
AutoScalingMinReplicas
integer
The minimum number of replicas for autoscaling. Defaults to 1.
HostPathMountPropagation
string
Share this volumes mount with other containers. If set to HostToContainer, this volume mount will receive all subsequent mounts that are mounted to this volume…
StorageInstanceName
object
1 property
TenantId
integer
The id of the tenant.
Preemptibility
string
Specifies whether the workload can be preempted by higher-priority workloads. Valid values are preemptible and non-preemptible. If explicitly set, this value t…
DistributedInferenceStartupPolicyFieldRules
object
1 property
DistributedInferenceServingPortAccess
object
5 properties
DistributedPolicyOverwriteRequestV2
object
2 properties
ImagePullSecrets
array
A list of references to Kubernetes secrets in the same namespace used for pulling container images.
ServingPortProtocol
string
The protocol used by the port, defaults to http
AutoScalingMetricFields
object
GitSecretRef
object
4 properties
ConfigMapsRules
object
2 properties
PreemptibilityOptions
object
1 property
Image
string
Docker image name. For more information, see [Images](https://kubernetes.io/docs/concepts/containers/images). The image name is mandatory for creating a worklo…
ReplicaDefaultsV2
object
32 properties
DistributedInferenceStartupPolicy
string
Determines when the worker pods should start during workload initialization. - LeaderCreated: Workers start after the leader pod is created. - LeaderReady: Wor…
CreateHomeDir
boolean
When set to true, creates a home directory for the container.
CpuCoreLimit
number
Limitations on the number of CPUs consumed by the workload (0.5, 1, .etc). The system guarantees that this workload will not be able to consume more than this…
RestartPolicyOptions
object
1 property
TolerationEffectOptions
object
1 property
GpuDevicesRequest
integer
Requested number of GPU devices. Currently if more than one device is requested, it is not possible to provide values for gpuMemory or gpuPortion.
InferencesMinReplicasField
object
1 property
ServingPortProtocolOptions
object
1 property
DistributedPolicyRulesV2
object
2 properties
AutoScalingRules
object
11 properties
NimServiceMultiNodeRules
object
1 property
PvcItems
array
Set of pvc persistent volume claims to use in the workload.
DataVolumeItems
array
Set of data volumes to use in the workload.
ReadOnlyRootFileSystem
boolean
If true, mounts the container's root filesystem as read-only.
ExposedUrls
array
Set of container ports that the workload exposes via URLs.
StringOption
object
One of the values that the field can be assigned to, when options police is applied on the field.
2 properties
1 required
WorkspacePolicyOverwriteRequestV2
object
2 properties
NodeSelectorTerm
object
A null or empty node selector term matches no objects. The requirements of them are ANDed.
1 property
ExposedUrlsDefaults
object
2 properties
PolicySyncStatusOfCluster
object
3 properties
2 required
InferencePolicyDefaultsV2
IntegerRules
object
7 properties
Tolerations
array
Set of tolerations to apply to the workload.
DistributedInferenceRestartPolicyOptions
object
1 property
Nfs
object
4 properties
ServingPortPort
integer
The port that the container running the inference service exposes (mandatory).
DistributedInferenceServingPortRules-2
object
Defines the configuration for the inference serving endpoint. This determines how applications or services can send inference requests to the workload.
GitAuthenticationMethod
string
Specifies the authentication method to use when accessing the Git repository. This is required for private repositories - password - Authenticate using a usern…
AutoScaling
object
DistributedPolicyDefaultsV2
object
2 properties
TerminateAfterPreemption
boolean
Indicates if the job should be terminated by the system after it has been preempted.
UidGidSourceOptions
object
1 property
PortServiceType
string
The service type of the port (mandatory).
ConcurrencyHardLimitField
object
1 property
RelatedUrl
object
A URL that is related to the workload. For example, a URL to an external server providing statistics or logging about the workload.
4 properties
NimCache
object
The specification of a NIM cache volume.
2 properties
GpuPortionLimit
number
Limitations on the portion consumed by the workload, per GPU device. The system guarantees The gpuPotionLimit must be no less than the gpuPortionRequest.
GpuRequestOptions
object
1 property
NimServiceServingPort
object
A port for accessing the inference service
7 properties
MpiLauncherCreationPolicy
string
Define whether the MPI Launcher is created in parallel with the workers, or if its creation is postponed until all workers are in Ready state. This prevents fa…
LabelsDefaults
object
1 property
S3Instance
object
ExposedUrlsRules
object
2 properties
ReplicaRulesV2
object
32 properties
DistributedInferenceServingPortRules
object
7 properties
ExtendedResources
array
Extended resources and their quantity.
PodAffinityTypeRules
object
TolerationRules
object
Toleration details.
6 properties
DistributedInferencePolicyDefaultsAndRulesV2
object
4 properties
Secret5
RelatedUrlsRules
object
2 properties
NimServiceDefaults
DistributedFrameworkRules
object
AuthorizationType
string
Specifies who can access the connection URL: - authenticatedUsers: Any authenticated user or service account can access the URL; the authorizedUsers and author…
DistributedImposedAssets
object
2 properties
SourceOfRule
object
This field is used by the system along with effective rules, in order to specify the org unit from which this effective rule has been derived. It should be lef…
5 properties
1 required
PortsDefaults
object
2 properties
PolicyReadOnlyFields
object
7 properties
5 required
PvcsDefaults
object
2 properties
InferencePolicyOverwriteRequestV2
object
2 properties
ConfigMapRules
object
4 properties
UidGidSource
string
Indicate the way to determine the user and group ids of the container. The options are a. fromTheImage - user and group ids are determined by the docker image…
SecretInstance2
object
InitializationTimeoutField
object
1 property
MetricThresholdField
object
1 property
InferencePolicyRulesV2
PvcRules
object
4 properties
TolerationOperatorOptions
object
1 property
S3Rules
object
3 properties
CpuMemoryLimit
string
Limitations on the CPU memory to allocate for this workload (1G, 20M, .etc). The system guarantees that this workload will not be able to consume more than thi…
GitInstance
object
ServingConfigurationRules
object
2 properties
EmptyDirsRules
object
2 properties
Pvc
GpuMemoryRequest
string
Required if and only if gpuRequestType is memory. States the GPU memory to allocate for the created workload, per GPU device. Note that the workload will not b…
NimServicePvcFields
object
4 properties
SecretFieldsUpdatable
object
2 properties
EmptyDirRules
object
3 properties
LargeShmRequest
boolean
A large /dev/shm device to mount into a container running the created workload. An shm is a shared file system mounted on RAM.
DistributedInferenceWorkersField
object
1 property
EnvironmentVariableSecret
object
Details of the secret and key use to populate the environment variable
2 properties
MinReplicas
integer
the lower limit for the number of worker pods to which the training job can scale down. (applicable only for PyTorch)
TrainingPolicyDefaultsAndRulesV2
object
4 properties
AutoScalingMetricOptions
object
1 property
HostPathMountPropagationOptions
object
1 property
RelatedUrlRules
object
Rules for the attributes of relatedUrls
3 properties
PvcAccessModesRules
object
Rules for access mode(s) for a newly created PVC.
3 properties
MpiLauncherCreationPolicyOptions
object
1 property
MatchExpressionOperator
string
Represents a key's relationship to a set of values (mandatory).
AllowPrivilegeEscalation
boolean
Allow the container running the workload and all launched processes to gain additional privileges after the workload starts. For more information consult the U…
SlotsPerWorker
integer
Specifies the number of slots per worker used in hostfile. Defaults to 1. (applicable only for MPI)
PolicySyncStatus
object
1 property
NumWorkers
integer
the number of workers that will be allocated for running the workload.
ArbitraryRules
object
3 properties
DefaultFromRule
object
2 properties
SupplementalGroups
string
Comma separated list of groups that the user running the container belongs to, in addition to the group indicated by runAsGid. Use only when the source uid/gid…
Completions
integer
Used with Hyperparameter Optimization. Specifies the number of successful pods the job should reach to be completed. The Job will be marked as successful once…
NimServiceWorkers
integer
Specifies the number of worker nodes to use when running the NIM service in multi-node.
Annotation
object
Annotation details to be populated into the container.
3 properties
GpuMemoryLimit
string
Limitation on the memory consumed by the workload, per GPU device. The system guarantees The gpuMemoryLimit must be no less than gpuMemoryRequest.
EnvironmentVariableConfigMap
object
Details of the configMap and key use to populate the environment variable
2 properties
FieldsToReset
array
set of fields in jsonpath format that is requested to clear their policy (default and rules)
DistributedInferenceServingPortAccessAuthorizationTypeEnum
string
Specifies who can send inference requests to the serving endpoint: Possible values: - public: No authorization is required. (Default) - authenticatedUsers: Any…
ServingPortRules
object
6 properties
TolerationEffectRules
object
NfsInstance
object
PodAffinity
object
Pod affinity scheduling rules (e.g. co-locate this workload in the same node, zone, etc. as some other workloads).
2 properties
AutoScalingMaxReplicas
integer
The maximum number of replicas for autoscaling. Defaults to minReplicas. Must be no less than minReplicas.
ConfigMapInstance
object
SecretsRules
object
2 properties
DistributedFrameworkOptions
object
1 property
ExtendedResourcesDefaults
object
2 properties
ServingPortProtocolRules
object
DistributedInferencePolicyV2
object
4 properties
RunAsNonRoot
boolean
Force the container to run as a non-root user.
NimServiceReplicas
integer
The number of replicas to deploy.
ServingPortMetricsPort
integer
The port where metrics are exposed, required only if it's different than the main port.
ItemRules
object
3 properties
AutoScalingScaleWindowSeconds
integer
The time window for autoscaling decisions, in seconds. Defaults to 300 seconds.
PriorityClass
string
Specifies the priority class for the workload, which determines its scheduling behavior. Valid values are: very-low, low, medium-low, medium, medium-high, high…
SshAuthMountPath
string
Specifies the directory where SSH keys are mounted. (applicable only for MPI)
PvcVolumeMode
string
Default volume mode for the PVC. Choose between Filesystem (default) or Block.
PolicyType
string
The type of policy which determines the kind of workloads it is affecting
Label
object
Label details to be populated into the container.
3 properties
GitItems
array
Set of git repositories to use in the workload.
SeccompProfileType
string
Indicates which kind of seccomp profile will be applied to the container. The options are a. RuntimeDefault - the container runtime default profile should be u…
DataVolumesRules
object
2 properties
ExposedUrlRules
object
Rules for the attributed of exposedUrls
7 properties
DistributedInferenceReplicasFieldRules
object
1 property
TrainingPolicyOverwriteRequestV2
object
2 properties
CpuMemoryRequest
string
The amount of CPU memory to allocate for this workload (1G, 20M, .etc). The workload will receive at least this amount of memory. Note that the workload will n…
DataVolumeRules
object
2 properties
DistributedInferencePolicyChangeRequestV2
object
3 properties
NfsRules
object
4 properties
ProbeHandlerRules
object
1 property
GpuPortionRequest
number
Required if and only if gpuRequestType is portion. States the portion of the GPU to allocate for the created workload, per GPU device, between 0 and 1. The def…
ImagePullSecretsRules
object
2 properties
DistributedInferenceWorkerDefaults
object
1 property
HostPathsRules
object
2 properties
HostPathInstance
object
InferencePolicyV2
object
4 properties
QuantityRules
object
6 properties
S3Items
array
Set of s3 buckets to use in the workload.
MaxReplicas
integer
the upper limit for the number of worker pods that can be set by the autoscaler. Cannot be smaller than MinReplicas. (applicable only for PyTorch)
ConfigMapsDefaults
object
2 properties
ConfigMap
object
4 properties
DistributedInferenceLeaderWorkerDefaultsV2
object
18 properties
NfssRules
object
2 properties
DataVolumesDefaults
object
2 properties
ServingPortExposedProtocolRules
object
ScaleToZeroRetentionField
object
1 property
BooleanRules
object
3 properties
ImagePullSecret
object
A reference to a secret in the same namespace used to pull container images.
3 properties
TolerationOperator
string
A key's relationship to the value. Equal uses key and value. Exists is equivalent to wildcard for value, so that a workload can tolerate all taints of a partic…
DepartmentIdOptional
string
The id of the department.
HostPathMountPropagationRules
object
PvcAccessModes
object
Default access mode(s) applied to newly created PVCs unless explicitly overridden.
3 properties
Specification
The full machine-readable OpenAPI contract behind this narrative.
Source
More from NVIDIA Run:ai 12
Other APIs NVIDIA Run:ai publishes across the network.
NVIDIA Run:ai Access Keys API
NVIDIA Run:ai Access rules API
NVIDIA Run:ai Administrator Command Line Interface API
NVIDIA Run:ai AI Applications API
NVIDIA Run:ai Applications API
NVIDIA Run:ai AuditLogs API
NVIDIA Run:ai Clusters API
NVIDIA Run:ai Compute API
NVIDIA Run:ai ConfigMap API
NVIDIA Run:ai Credentials API
NVIDIA Run:ai Datasources API
NVIDIA Run:ai Datavolumes API
#Policy