> ## Documentation Index > Fetch the complete documentation index at: https://docs.nebius.com/llms.txt > Use this file to discover all available pages before exploring further. # Create > Creates an endpoint. ## OpenAPI ````yaml https://api.nebius.cloud/openapi.json post /ai/v1/endpoints openapi: 3.0.3 info: title: Nebius API version: version not set servers: - url: https://api.nebius.cloud security: - bearerAuth: [] tags: - name: nebius.ai.v1.EndpointService - name: nebius.ai.v1.JobService - name: nebius.applications.v1alpha1.K8sReleaseService - name: nebius.audit.v2.AuditEventExportService - name: nebius.audit.v2.AuditEventService - name: nebius.billing.v1.CalculatorService - name: nebius.billing.v1alpha1.CalculatorService - name: nebius.billing.v1alpha1.OneTimeExportService - name: nebius.capacity.v1.CapacityAllowanceService - name: nebius.capacity.v1.CapacityBlockGroupService - name: nebius.capacity.v1.CapacityIntervalService - name: nebius.capacity.v1.ResourceAdviceService - name: nebius.compute.v1.DiskService - name: nebius.compute.v1.DiskSnapshotService - name: nebius.compute.v1.FilesystemService - name: nebius.compute.v1.GpuClusterService - name: nebius.compute.v1.ImageService - name: nebius.compute.v1.InstanceService - name: nebius.compute.v1.MaintenanceService - name: nebius.compute.v1.NVLInstanceGroupService - name: nebius.compute.v1.NodeService - name: nebius.compute.v1.PlatformService - name: nebius.dns.v1.RecordService - name: nebius.dns.v1.ZoneService - name: nebius.iam.v1.AccessKeyService - name: nebius.iam.v1.AccessPermitService - name: nebius.iam.v1.AuthPublicKeyService - name: nebius.iam.v1.FederatedCredentialsService - name: nebius.iam.v1.FederationCertificateService - name: nebius.iam.v1.FederationService - name: nebius.iam.v1.GroupMembershipService - name: nebius.iam.v1.GroupService - name: nebius.iam.v1.InvitationService - name: nebius.iam.v1.ProfileService - name: nebius.iam.v1.ProjectService - name: nebius.iam.v1.ServiceAccountService - name: nebius.iam.v1.SessionManagementService - name: nebius.iam.v1.StaticKeyService - name: nebius.iam.v1.TenantService - name: nebius.iam.v1.TenantUserAccountService - name: nebius.iam.v1.TenantUserAccountWithAttributesService - name: nebius.iam.v2.AccessKeyService - name: nebius.iam.v2.ProjectService - name: nebius.iam.v2.TenantService - name: nebius.kms.v1.AsymmetricCryptoService - name: nebius.kms.v1.AsymmetricKeyService - name: nebius.kms.v1.SymmetricCryptoService - name: nebius.kms.v1.SymmetricKeyService - name: nebius.logging.agentmanager.v1.VersionService - name: nebius.logging.v1.LogExportService - name: nebius.maintenance.v1alpha1.MaintenanceService - name: nebius.mk8s.v1.ClusterService - name: nebius.mk8s.v1.NodeGroupService - name: nebius.mk8s.v1alpha1.ClusterService - name: nebius.mk8s.v1alpha1.NodeGroupService - name: nebius.msp.mlflow.v1alpha1.ClusterService - name: nebius.msp.postgresql.v1alpha1.BackupService - name: nebius.msp.postgresql.v1alpha1.ClusterService - name: nebius.mysterybox.v1.PayloadService - name: nebius.mysterybox.v1.SecretService - name: nebius.mysterybox.v1.SecretVersionService - name: nebius.quotas.v1.QuotaAllowanceService - name: nebius.registry.v1.ArtifactService - name: nebius.registry.v1.RegistryService - name: nebius.storage.v1.BucketService - name: nebius.storage.v1.TransferService - name: nebius.storage.v1alpha1.TransferService - name: nebius.tunnel.v1.TunnelService - name: nebius.vpc.v1.AllocationService - name: nebius.vpc.v1.NetworkService - name: nebius.vpc.v1.PoolService - name: nebius.vpc.v1.RouteService - name: nebius.vpc.v1.RouteTableService - name: nebius.vpc.v1.SecurityGroupService - name: nebius.vpc.v1.SecurityRuleService - name: nebius.vpc.v1.SubnetService - name: nebius.vpc.v1.TargetGroupService - name: nebius.vpc.v1alpha1.AllocationService - name: nebius.vpc.v1alpha1.NetworkService - name: nebius.vpc.v1alpha1.PoolService - name: nebius.vpc.v1alpha1.ScopeService - name: nebius.vpc.v1alpha1.SubnetService paths: /ai/v1/endpoints: post: tags: - nebius.ai.v1.EndpointService summary: Create description: Creates an endpoint. operationId: EndpointService_Create requestBody: content: application/json: schema: $ref: '#/components/schemas/v1CreateEndpointRequest' required: true x-originalParamName: body responses: '200': content: application/json: schema: $ref: '#/components/schemas/commonV1Operation' description: A successful response. components: schemas: v1CreateEndpointRequest: properties: dryRun: description: Dry run creation of the endpoint. type: boolean metadata: $ref: '#/components/schemas/commonV1ResourceMetadata' spec: $ref: '#/components/schemas/v1EndpointSpec' required: - metadata - spec type: object commonV1Operation: properties: createdAt: description: Creation timestamp. format: date-time type: string createdBy: description: ID of the user or service account who initiated the operation. type: string description: description: Human-readable description of the operation. 0-256 characters long. type: string finishedAt: description: The time when the operation has finished. format: date-time type: string id: description: ID of the operation. type: string progressData: $ref: '#/components/schemas/protobufAny' progressTracker: $ref: '#/components/schemas/v1ProgressTracker' request: $ref: '#/components/schemas/protobufAny' requestHeaders: additionalProperties: $ref: '#/components/schemas/OperationRequestHeader' title: >- The request headers that are essential for the request that generated the operation. For instance, `x-resetmask`. Without these headers the request might have been processed differently if repeated. All the header names *must* be converted to lower case. Validator is based on: https://httpwg.org/specs/rfc9110.html#considerations.for.new.field.names type: object resourceId: description: >- ID of the resource that this operation creates, updates, deletes or otherwise changes. If the operation affects multiple resources or does not affect any API resources at all (e.g. a routine maintenance operation visible to the user), the [resource_id] must be empty. type: string status: $ref: '#/components/schemas/googleRpcStatus' type: object commonV1ResourceMetadata: description: Common resource metadata. properties: createdAt: description: Timestamp indicating when the resource was created. format: date-time readOnly: true type: string id: description: Identifier for the resource, unique for its resource type. type: string labels: additionalProperties: type: string description: Labels associated with the resource. type: object name: description: Human readable name for the resource. type: string parentId: description: Identifier of the parent resource to which the resource belongs. type: string resourceVersion: description: >- Version of the resource for safe concurrent modifications and consistent reads. Positive and monotonically increases on each resource spec change (but *not* on each change of the resource's container(s) or status). Service allows zero value or current. format: int64 type: string updatedAt: description: Timestamp indicating when the resource was last updated. format: date-time readOnly: true type: string required: - parentId type: object v1EndpointSpec: description: EndpointSpec defines a endpoint that will be run. properties: args: description: The arguments to pass to the entrypoint command. type: string authToken: description: >- Authentication token needed to access the endpoint. Authentication can only be enabled if the endpoint exposes one and only one HTTP port. Mutually exclusive with `auth_token_mysterybox_secret`. If not provided, authentication will be disabled. type: string authTokenMysteryboxSecret: $ref: '#/components/schemas/v1EndpointSpecMysteryBoxSecretRef' containerCommand: description: The entrypoint command for the endpoint's container. type: string disk: $ref: '#/components/schemas/v1EndpointSpecDiskSpec' environmentVariables: description: Specifies the environment variables for the endpoint's container. items: $ref: '#/components/schemas/v1EndpointSpecEnvironmentVariable' type: array image: description: The Docker image to use for the endpoint's container. type: string injectedFiles: description: >- Small config files injected into the container before the user process starts. Intended for configs, not datasets. items: $ref: '#/components/schemas/v1EndpointSpecFileInjection' type: array platform: description: Compute platform that the endpoint will be run on. type: string ports: description: Specifies the ports that the endpoint exposes. items: $ref: '#/components/schemas/v1EndpointSpecPort' type: array preemptible: description: >- Whether to use a preemptible VM for the endpoint. Preemptible VMs are cheaper but can be stopped by the platform at any time. type: boolean preset: description: Compute preset that the endpoint will be run on. type: string publicIp: description: Whether to assign a public IP to the endpoint. type: boolean registryCredentials: $ref: '#/components/schemas/v1EndpointSpecRegistryCredentials' shmSizeBytes: description: Shared memory size in bytes for the endpoint's container. format: int64 type: string sshAuthorizedKeys: description: Public keys to be authorized for SSH access to the job. items: type: string type: array subnetId: description: Subnet ID where the endpoint will be deployed. type: string volumes: description: Volumes to be mounted into the endpoint's container. items: $ref: '#/components/schemas/v1EndpointSpecVolumeMount' type: array workingDir: description: The working directory for the endpoint's container. type: string required: - image - platform - preset - disk - subnetId type: object protobufAny: additionalProperties: {} description: >- `Any` contains an arbitrary serialized protocol buffer message along with a URL that describes the type of the serialized message properties: '@type': description: >- A URL/resource name that uniquely identifies the type of the serialized protocol buffer message. This string must contain at least one "/" character. The last segment of the URL's path must represent the fully qualified name of the type (as in `path/google.protobuf.Duration`). The name should be in a canonical form (e.g., leading "." is not accepted). In practice, teams usually precompile into the binary all types that they expect it to use in the context of Any. However, for URLs which use the scheme `http`, `https`, or no scheme, one can optionally set up a type server that maps type URLs to message definitions as follows: * If no scheme is provided, `https` is assumed. * An HTTP GET on the URL must yield a [google.protobuf.Type][] value in binary format, or produce an error. * Applications are allowed to cache lookup results based on the URL, or have them precompiled into a binary to avoid any lookup. Therefore, binary compatibility needs to be preserved on changes to types. (Use versioned type names to manage breaking changes.) Note: this functionality is not currently available in the official protobuf release, and it is not used for type URLs beginning with type.googleapis.com. As of May 2023, there are no widely used type server implementations and no plans to implement one. Schemes other than `http`, `https` (or the empty scheme) might be used with implementation specific semantics. type: string type: object v1ProgressTracker: description: >- Information for tracking the progress of a task (e.g., an API operation or background maintenance action). This serves as a UI contract and may be directly reflected in CLI, Console, and other tools. properties: description: description: >- Human-readable description of the currently executing task, e.g., "Downloading dat-groot-llm.tar.xz". Relates to the overall task, not individual steps. SHOULD match the Operation's `description` field when this ProgressTracker tracks an API Operation's progress. MAY provide more detailed overall information. MUST be suitable for display to public users. Even if the task is internal, this description may be shown to end users in overall task step descriptions. For individual step descriptions, use `Step.description`. type: string estimatedFinishedAt: description: |- Estimated completion timestamp for the task. MUST be absent if the task is completed (i.e., finished_at is set). MAY be absent if the estimate is unknown. MAY be updated as the progress estimate changes. format: date-time type: string finishedAt: description: >- Completion timestamp of the task. MUST be absent for running tasks; MUST be present for completed tasks. MUST match the operation's finished_at timestamp when tracking an API Operation's progress. format: date-time type: string startedAt: description: >- Timestamp when the overall task started. MUST match the operation's created_at timestamp when tracking an API Operation's progress. format: date-time type: string steps: description: >- Detailed information about the task steps that are currently running or have been finished. - For a completed task, the step list MUST either be empty or contain finished steps only. - For a running task, the step list MAY be empty if the task consists of a single step or no detailed information is available. A non-empty step list MUST include all currently running steps and MAY include some or all finished steps. - Service SHOULD provide both running and finished steps if the task has around 10-20 steps, and provide only running steps otherwise. Service MUST consistently choose either the "running steps only" or the "running + finished steps" behavior. - The UI, CLI etc. MAY choose not to display all steps provided by the service. items: $ref: '#/components/schemas/ProgressTrackerStep' type: array workDone: $ref: '#/components/schemas/ProgressTrackerWorkDone' type: object OperationRequestHeader: properties: values: items: type: string title: The values of a particular header from a request type: array title: >- Request header is a container for all the values of a particular header of a request because there is no such thing as `map` type: object googleRpcStatus: description: >- The `Status` type defines a logical error model that is suitable for different programming environments, including REST APIs and RPC APIs. It is used by [gRPC](https://github.com/grpc). Each `Status` message contains three pieces of data: error code, error message, and error details. You can find out more about this error model and how to work with it in the [API Design Guide](https://cloud.google.com/apis/design/errors). properties: code: description: |- The status code, which should be an enum value of [google.rpc.Code][google.rpc.Code]. format: int32 type: integer details: description: >- A list of messages that carry the error details. There is a common set of message types for APIs to use. items: $ref: '#/components/schemas/protobufAny' type: array message: description: >- A developer-facing error message, which should be in English. Any user-facing error message should be localized and sent in the [google.rpc.Status.details][google.rpc.Status.details] field, or localized by the client. type: string type: object v1EndpointSpecMysteryBoxSecretRef: description: Reference to a MysteryBox secret. properties: secretId: description: MysteryBox secret ID. type: string versionId: description: MysteryBox secret version ID. type: string type: object v1EndpointSpecDiskSpec: properties: sizeBytes: description: Disk size in bytes. format: int64 type: string type: $ref: '#/components/schemas/nebiusComputeV1DiskSpecDiskType' required: - type type: object v1EndpointSpecEnvironmentVariable: description: >- EnvironmentVariable defines an environment variable for the endpoint's container. properties: mysteryboxSecret: $ref: '#/components/schemas/v1EndpointSpecMysteryBoxSecretRef' name: description: The name of the environment variable. type: string value: description: |- Environment variable value. Mutually exclusive with `mysterybox_secret`. type: string required: - name type: object v1EndpointSpecFileInjection: description: FileInjection materializes a small file inside the container at launch. properties: containerPath: description: |- Absolute path inside the container where the content is written. Must be a clean absolute path: root, trailing slashes, empty path segments, "." and ".." are not allowed. type: string content: description: >- File content. Between 1 byte and 64 KiB (one mystery box secret payload). Not returned by read methods. format: byte type: string required: - containerPath type: object v1EndpointSpecPort: properties: containerPort: description: Container port. format: int32 type: integer hostPort: description: |- Host port. If not specified, will be same as container_port. format: int32 type: integer protocol: $ref: '#/components/schemas/v1EndpointSpecPortProtocol' required: - protocol type: object v1EndpointSpecRegistryCredentials: properties: mysteryboxSecretVersion: description: |- Secret version storing the registry credentials. Must have keys "REGISTRY_USERNAME" and "REGISTRY_PASSWORD". type: string password: description: Registry password for private Docker registry. type: string username: description: Registry username for private Docker registry. type: string type: object v1EndpointSpecVolumeMount: description: VolumeMount represents a volume mount for the endpoint's container. properties: containerPath: description: |- Path inside the endpoint's container where the volume is mounted. Must be an absolute path. type: string mode: $ref: '#/components/schemas/v1EndpointSpecVolumeMountMode' s3Config: $ref: '#/components/schemas/v1EndpointSpecVolumeMountS3Config' source: description: >- Source of the volume mount. Can be a name or an ID of Nebius Storage bucket or filesystem, or an S3 URI (e.g. "s3://bucket-name") when using external S3 storage. type: string sourcePath: description: |- Path inside the source volume. Optional. type: string required: - mode type: object ProgressTrackerStep: description: |- Represents a basic step in the task. Fields are binary-compatible with ProgressTracker for easier processing. properties: description: description: >- Human-readable description of the step, e.g., "Connecting to localhost:8080". MUST be suitable for display to public users. Private descriptions must be filtered server-side based on the use case. type: string finishedAt: description: >- Timestamp when the step finished. MUST be absent for running steps; MUST be present for completed steps. format: date-time type: string startedAt: description: Timestamp when the step started. format: date-time type: string workDone: $ref: '#/components/schemas/ProgressTrackerWorkDone' type: object ProgressTrackerWorkDone: description: >- Information about the work done by the task or its step, expressed in ticks (abstract work units). Each tick may represent a real measurement (e.g., VMs created, KiB uploaded), number of substeps or a percentage of work completed. properties: doneTickCount: description: >- Number of ticks completed so far. MUST be between 0 and total_tick_count, inclusive. MUST equal total_tick_count if the task or step is finished. format: int64 type: string totalTickCount: description: >- Total number of ticks (work units) to be completed. MUST be greater than 0. format: int64 type: string type: object nebiusComputeV1DiskSpecDiskType: default: UNSPECIFIED enum: - UNSPECIFIED - NETWORK_SSD - NETWORK_HDD - NETWORK_SSD_NON_REPLICATED - NETWORK_SSD_IO_M3 title: >- the list of available types will be clarified later, it is not final version type: string v1EndpointSpecPortProtocol: default: PROTOCOL_UNSPECIFIED description: |- Represents protocol of the endpoint's port which will be exposed. - HTTP: HTTP protocol. - TCP: TCP protocol. - UDP: UDP protocol. enum: - PROTOCOL_UNSPECIFIED - HTTP - TCP - UDP type: string v1EndpointSpecVolumeMountMode: default: MODE_UNSPECIFIED description: |- Mode that will be used to mount the volume. - READ_ONLY: Read-only mode. - READ_WRITE: Read-write mode. enum: - MODE_UNSPECIFIED - READ_ONLY - READ_WRITE type: string v1EndpointSpecVolumeMountS3Config: description: |- Config for accessing an external S3-compatible storage. The bucket name is specified in the `source` field as an S3 URI. properties: credentials: $ref: '#/components/schemas/v1EndpointSpecVolumeMountS3ConfigS3Credentials' endpoint: description: S3-compatible endpoint URL (e.g. "https://s3.amazonaws.com"). type: string mysteryboxSecret: $ref: >- #/components/schemas/v1EndpointSpecVolumeMountS3ConfigMysteryBoxSecretRef region: description: S3 region. type: string required: - endpoint - region type: object v1EndpointSpecVolumeMountS3ConfigS3Credentials: description: Inline S3 credentials. properties: accessKeyId: description: Access key ID. type: string secretAccessKey: description: Secret access key. type: string sessionToken: description: Session token (optional, for temporary credentials). type: string required: - accessKeyId - secretAccessKey type: object v1EndpointSpecVolumeMountS3ConfigMysteryBoxSecretRef: description: Reference to a MysteryBox secret. properties: secretId: description: MysteryBox secret ID. type: string versionId: description: MysteryBox secret version ID. type: string type: object securitySchemes: bearerAuth: description: The Authorization header expects a Bearer token. scheme: bearer type: http ````