UNPKG

@pulumi/digitalocean

Version:

A Pulumi package for creating and managing DigitalOcean cloud resources.

229 lines • 8.75 kB
import * as pulumi from "@pulumi/pulumi"; import * as inputs from "./types/input"; import * as outputs from "./types/output"; /** * Provides a DigitalOcean Dedicated Inference resource. This can be used to create, * modify, and delete dedicated inference endpoints for running GPU-accelerated * model inference. * * ## Example Usage * * ```typescript * import * as pulumi from "@pulumi/pulumi"; * import * as digitalocean from "@pulumi/digitalocean"; * * const example = new digitalocean.DedicatedInference("example", { * name: "my-inference-endpoint", * region: "tor1", * modelDeployments: [{ * modelSlug: "deepseek-r1-distill-qwen-14b", * modelProvider: "digitalocean", * accelerators: [{ * acceleratorSlug: "gpu-h100x1-80gb", * scale: 1, * type: "nvidia_h100", * }], * }], * }); * ``` * * ### With Public Endpoint * * ```typescript * import * as pulumi from "@pulumi/pulumi"; * import * as digitalocean from "@pulumi/digitalocean"; * * const _public = new digitalocean.DedicatedInference("public", { * name: "my-public-inference", * region: "tor1", * enablePublicEndpoint: true, * modelDeployments: [{ * modelSlug: "deepseek-r1-distill-qwen-14b", * modelProvider: "digitalocean", * accelerators: [{ * acceleratorSlug: "gpu-h100x1-80gb", * scale: 1, * type: "nvidia_h100", * }], * }], * }); * ``` * * ### With VPC * * ```typescript * import * as pulumi from "@pulumi/pulumi"; * import * as digitalocean from "@pulumi/digitalocean"; * * const _private = new digitalocean.DedicatedInference("private", { * name: "my-private-inference", * region: "tor1", * vpcUuid: example.id, * modelDeployments: [{ * modelSlug: "deepseek-r1-distill-qwen-14b", * modelProvider: "digitalocean", * accelerators: [{ * acceleratorSlug: "gpu-h100x1-80gb", * scale: 1, * type: "nvidia_h100", * }], * }], * }); * ``` * * ## Import * * Dedicated inference endpoints can be imported using their `id`, e.g. * * ```sh * $ pulumi import digitalocean:index/dedicatedInference:DedicatedInference example endpoint-id * ``` */ export declare class DedicatedInference extends pulumi.CustomResource { /** * Get an existing DedicatedInference resource's state with the given name, ID, and optional extra * properties used to qualify the lookup. * * @param name The _unique_ name of the resulting resource. * @param id The _unique_ provider ID of the resource to lookup. * @param state Any extra arguments used during the lookup. * @param opts Optional settings to control the behavior of the CustomResource. */ static get(name: string, id: pulumi.Input<pulumi.ID>, state?: DedicatedInferenceState, opts?: pulumi.CustomResourceOptions): DedicatedInference; /** * Returns true if the given object is an instance of DedicatedInference. This is designed to work even * when multiple copies of the Pulumi SDK have been loaded into the same process. */ static isInstance(obj: any): obj is DedicatedInference; /** * The date and time when the dedicated inference endpoint was created. */ readonly createdAt: pulumi.Output<string>; /** * Whether to enable a public HTTPS endpoint for the dedicated inference endpoint. Defaults to `false`. This field is immutable after creation and changing it forces a new resource. */ readonly enablePublicEndpoint: pulumi.Output<boolean | undefined>; /** * A HuggingFace token for accessing gated models. */ readonly huggingFaceToken: pulumi.Output<string | undefined>; /** * The list of model deployments to run on the dedicated inference endpoint. Each `modelDeployments` block supports: */ readonly modelDeployments: pulumi.Output<outputs.DedicatedInferenceModelDeployment[]>; /** * A human-readable name for the dedicated inference endpoint. */ readonly name: pulumi.Output<string>; /** * The fully-qualified domain name of the private endpoint. */ readonly privateEndpointFqdn: pulumi.Output<string>; /** * The fully-qualified domain name of the public endpoint, if enabled. */ readonly publicEndpointFqdn: pulumi.Output<string>; /** * The region slug where the dedicated inference endpoint will be deployed. Changing this forces a new resource. */ readonly region: pulumi.Output<string>; /** * The current status of the dedicated inference endpoint. */ readonly status: pulumi.Output<string>; /** * The date and time when the dedicated inference endpoint was last updated. */ readonly updatedAt: pulumi.Output<string>; /** * The UUID of the VPC to deploy the dedicated inference endpoint into. Changing this forces a new resource. */ readonly vpcUuid: pulumi.Output<string | undefined>; /** * Create a DedicatedInference resource with the given unique name, arguments, and options. * * @param name The _unique_ name of the resource. * @param args The arguments to use to populate this resource's properties. * @param opts A bag of options that control this resource's behavior. */ constructor(name: string, args: DedicatedInferenceArgs, opts?: pulumi.CustomResourceOptions); } /** * Input properties used for looking up and filtering DedicatedInference resources. */ export interface DedicatedInferenceState { /** * The date and time when the dedicated inference endpoint was created. */ createdAt?: pulumi.Input<string | undefined>; /** * Whether to enable a public HTTPS endpoint for the dedicated inference endpoint. Defaults to `false`. This field is immutable after creation and changing it forces a new resource. */ enablePublicEndpoint?: pulumi.Input<boolean | undefined>; /** * A HuggingFace token for accessing gated models. */ huggingFaceToken?: pulumi.Input<string | undefined>; /** * The list of model deployments to run on the dedicated inference endpoint. Each `modelDeployments` block supports: */ modelDeployments?: pulumi.Input<pulumi.Input<inputs.DedicatedInferenceModelDeployment>[] | undefined>; /** * A human-readable name for the dedicated inference endpoint. */ name?: pulumi.Input<string | undefined>; /** * The fully-qualified domain name of the private endpoint. */ privateEndpointFqdn?: pulumi.Input<string | undefined>; /** * The fully-qualified domain name of the public endpoint, if enabled. */ publicEndpointFqdn?: pulumi.Input<string | undefined>; /** * The region slug where the dedicated inference endpoint will be deployed. Changing this forces a new resource. */ region?: pulumi.Input<string | undefined>; /** * The current status of the dedicated inference endpoint. */ status?: pulumi.Input<string | undefined>; /** * The date and time when the dedicated inference endpoint was last updated. */ updatedAt?: pulumi.Input<string | undefined>; /** * The UUID of the VPC to deploy the dedicated inference endpoint into. Changing this forces a new resource. */ vpcUuid?: pulumi.Input<string | undefined>; } /** * The set of arguments for constructing a DedicatedInference resource. */ export interface DedicatedInferenceArgs { /** * Whether to enable a public HTTPS endpoint for the dedicated inference endpoint. Defaults to `false`. This field is immutable after creation and changing it forces a new resource. */ enablePublicEndpoint?: pulumi.Input<boolean | undefined>; /** * A HuggingFace token for accessing gated models. */ huggingFaceToken?: pulumi.Input<string | undefined>; /** * The list of model deployments to run on the dedicated inference endpoint. Each `modelDeployments` block supports: */ modelDeployments: pulumi.Input<pulumi.Input<inputs.DedicatedInferenceModelDeployment>[]>; /** * A human-readable name for the dedicated inference endpoint. */ name?: pulumi.Input<string | undefined>; /** * The region slug where the dedicated inference endpoint will be deployed. Changing this forces a new resource. */ region: pulumi.Input<string>; /** * The UUID of the VPC to deploy the dedicated inference endpoint into. Changing this forces a new resource. */ vpcUuid?: pulumi.Input<string | undefined>; } //# sourceMappingURL=dedicatedInference.d.ts.map