> ## Documentation Index > Fetch the complete documentation index at: https://docs.coreweave.com/llms.txt > Use this file to discover all available pages before exploring further. # Dedicated Inference API reference > Authentication, protocols, status values, and error formats for the Dedicated Inference API The Dedicated Inference API, published as the CoreWeave Inference API, provides programmatic control over inference [gateways](/products/inference/gateways), model [deployments](/products/inference/models), [capacity claims](/products/inference/scaling#capacity-claims), and checkpoint hot-loads for RL Rollouts. The API is available at `api.coreweave.com`. This API manages Dedicated Inference resources. To call catalog models or LoRA adapters hosted by Serverless Inference, see the [Serverless Inference API reference](/products/inference/serverless/api-reference). To compare Serverless Inference, Dedicated Inference, and Inference on CKS, see [About CoreWeave Inference](/products/inference#choose-a-deployment-option). This page covers cross-cutting topics: authentication, protocols, status values, error formats, and the OpenAI-compatible inference endpoint. For per-endpoint request and response schemas, see the per-operation pages under each service in the left sidebar: * [CapacityClaimService](/products/inference/reference/api-overview/capacityclaimservice/list-capacity-claims) * [DeploymentService](/products/inference/reference/api-overview/deploymentservice/list-deployments) * [GatewayService](/products/inference/reference/api-overview/gatewayservice/list-gateways) The Dedicated Inference API is versioned as `v1alpha1`. APIs may change before general availability. ## Authentication The Dedicated Inference API uses bearer token authentication to identify the caller and authorize each request. All API requests must include a CoreWeave API access token in the `Authorization` header as a Bearer token. The token must belong to a user with the [Inference Viewer or Inference Admin role](/security/iam/access-policies), depending on the operation. Replace `[API-TOKEN]` with your CoreWeave API access token. ```bash theme={"system"} curl "https://api.coreweave.com/v1alpha1/inference/gateways" \ -H "Authorization: Bearer [API-TOKEN]" ``` For details on obtaining an API token, see [Manage API access tokens](/security/authn-authz/manage-api-access-tokens). ## Protocol support You can call the Dedicated Inference API over several transport protocols, depending on your client tooling and performance needs: | Protocol | Description | | - | - | | REST/JSON | Standard HTTP/1.1 with JSON request and response bodies. All examples in this documentation use REST. | | gRPC | Protocol buffers over HTTP/2 for high-performance programmatic access. | | Connect | gRPC-compatible protocol with improved browser and HTTP/1.1 support. | ## Schema and SDKs The Dedicated Inference API is defined in Protobuf and published to CoreWeave's public Buf Schema Registry, so you can install a generated client instead of writing HTTP calls by hand. * **Public BSR module**: [buf.build/coreweave/inference](https://buf.build/coreweave/inference) (module name `buf.build/coreweave/inference`). * **Generated SDKs**: [Inference SDKs on the Buf Schema Registry](https://buf.build/coreweave/inference/sdks/main%3Aconnectrpc). Clients are available for the Connect, gRPC, and Protobuf ecosystems across languages including Go, Python, TypeScript, Java, Kotlin, Rust, and Swift. * **Services**: `GatewayService`, `DeploymentService`, `CapacityClaimService`, and `HotLoadService` (package `coreweave.inference.v1alpha1`). Select an ecosystem and language on the Buf Schema Registry to get the install command for your package manager. Generated clients target the same API host and Bearer token authentication described in [Authentication](#authentication). Generated clients cover the Dedicated Inference API, which is what you use to create and manage gateways, deployments, capacity claims, and hot-loads. To send inference requests to a deployed model, use any OpenAI-compatible client against your gateway endpoint. See [OpenAI-compatible endpoint](#openai-compatible-endpoint). ## Query parameters List endpoints support the following query parameter: | Parameter | Type | Description | | - | - | - | | `updatedAfter` | `date-time` | Filter resources to those updated after the specified timestamp (ISO 8601 format). | ## Status values Use these status values to determine the lifecycle state of a resource when polling or reconciling state in your application. All resources share a common set of status values: | Status | Description | | - | - | | `STATUS_UNSPECIFIED` | Status has not been set. | | `STATUS_CREATING` | Resource is being provisioned. | | `STATUS_READY` | Resource is active and operational. | | `STATUS_UPDATING` | Resource configuration is being updated. | | `STATUS_DELETING` | Resource is being removed. | | `STATUS_ERROR` | Resource has an error that isn't terminal. Either CoreWeave temporarily couldn't check the resource's readiness, which clears on its own, or a deployment that runs only on reserved capacity has none allocated in its zone, which clears once capacity is allocated. Check `status.conditions` for the cause. | | `STATUS_FAILED` | Resource encountered a terminal error. | Each resource includes a `conditions` array in its status with detailed information about the current state, including timestamps, reasons, and human-readable messages. Each condition's `status` field is `STATUS_TRUE`, `STATUS_FALSE`, or `STATUS_UNSPECIFIED` when it has not been set. ## Error responses When a request fails, the API returns a structured error body that your client can parse to surface details to users or trigger retries. Error responses follow the standard format: ```json theme={"system"} { "code": 400, "message": "name is required", "details": [] } ``` ## OpenAI-compatible endpoint Deployed models expose an OpenAI-compatible completions endpoint through their associated gateway. The endpoint URL depends on the gateway's routing strategy. See [Gateways](/products/inference/gateways#routing-strategies) for details on how each routing strategy constructs the request URL. For a complete walkthrough, see the [Getting started](/products/inference/getting-started) guide. ## OpenAPI specification The complete OpenAPI 3.0 specification for the Dedicated Inference API is embedded below. Expand the code block by clicking "See all" at the bottom, or use the download and copy buttons in the title bar. ```yaml expandable title="Inference OpenAPI" theme={"system"} openapi: 3.0.3 info: title: CoreWeave Inference API version: 0.0.1 description: The CoreWeave Inference API provides programmatic control over inference gateways, model deployments, and capacity claims. servers: - url: https://api.coreweave.com description: CoreWeave production API. paths: /v1alpha1/inference/capacityclaims: get: operationId: CapacityClaimService_ListCapacityClaims summary: List capacity claims description: Returns every CapacityClaim that the caller can read. parameters: - name: updatedAfter in: query description: Filter by those updated after this timestamp schema: type: string format: date-time responses: '200': description: OK content: application/json: schema: $ref: '#/components/schemas/ListCapacityClaimsResponse' default: description: Default error response content: application/json: schema: $ref: '#/components/schemas/Status' tags: - CapacityClaimService post: operationId: CapacityClaimService_CreateCapacityClaim summary: Create capacity claim description: Creates a new CapacityClaim that reserves GPU hardware for inference deployments. requestBody: required: true content: application/json: schema: $ref: '#/components/schemas/CreateCapacityClaimRequest' responses: '200': description: OK content: application/json: schema: $ref: '#/components/schemas/CreateCapacityClaimResponse' default: description: Default error response content: application/json: schema: $ref: '#/components/schemas/Status' tags: - CapacityClaimService /v1alpha1/inference/capacityclaims/parameters: get: operationId: CapacityClaimService_GetCapacityClaimParameters summary: Get capacity claim parameters description: Returns the available instance types per Availability Zone for use when creating a CapacityClaim. responses: '200': description: OK content: application/json: schema: $ref: '#/components/schemas/GetCapacityClaimParametersResponse' default: description: Default error response content: application/json: schema: $ref: '#/components/schemas/Status' tags: - CapacityClaimService /v1alpha1/inference/capacityclaims/{id}: get: operationId: CapacityClaimService_GetCapacityClaim summary: Get capacity claim description: Retrieves a single CapacityClaim by its UUID. parameters: - name: id in: path description: The unique identifier of the CapacityClaim to get, UUID format required: true schema: type: string responses: '200': description: OK content: application/json: schema: $ref: '#/components/schemas/GetCapacityClaimResponse' default: description: Default error response content: application/json: schema: $ref: '#/components/schemas/Status' tags: - CapacityClaimService delete: operationId: CapacityClaimService_DeleteCapacityClaim summary: Delete capacity claim description: Deletes a CapacityClaim by ID. Returns the deleted CapacityClaim. parameters: - name: id in: path description: The unique identifier of the CapacityClaim to delete, UUID format required: true schema: type: string responses: '200': description: OK content: application/json: schema: $ref: '#/components/schemas/DeleteCapacityClaimResponse' default: description: Default error response content: application/json: schema: $ref: '#/components/schemas/Status' tags: - CapacityClaimService patch: operationId: CapacityClaimService_UpdateCapacityClaim summary: Update capacity claim description: Updates an existing CapacityClaim. The full `resources` block is required and replaces the existing reservation. parameters: - name: id in: path description: The unique identifier of the CapacityClaim to update, UUID format required: true schema: type: string requestBody: required: true content: application/json: schema: $ref: '#/components/schemas/UpdateCapacityClaimRequest' responses: '200': description: OK content: application/json: schema: $ref: '#/components/schemas/UpdateCapacityClaimResponse' default: description: Default error response content: application/json: schema: $ref: '#/components/schemas/Status' tags: - CapacityClaimService /v1alpha1/inference/deployments: get: operationId: DeploymentService_ListDeployments summary: List deployments description: Returns every deployment that the caller can read, optionally filtered by parent gateway or update timestamp. parameters: - name: parentGatewayId in: query description: The ID of the parent gateway to filter deployments by schema: type: string - name: updatedAfter in: query description: Filter by those updated after this timestamp schema: type: string format: date-time responses: '200': description: OK content: application/json: schema: $ref: '#/components/schemas/ListDeploymentsResponse' default: description: Default error response content: application/json: schema: $ref: '#/components/schemas/Status' tags: - DeploymentService post: operationId: DeploymentService_CreateDeployment summary: Create deployment description: Creates a new model deployment associated with one or more gateways. requestBody: required: true content: application/json: schema: $ref: '#/components/schemas/CreateDeploymentRequest' responses: '200': description: OK content: application/json: schema: $ref: '#/components/schemas/CreateDeploymentResponse' default: description: Default error response content: application/json: schema: $ref: '#/components/schemas/Status' tags: - DeploymentService /v1alpha1/inference/deployments/parameters: get: operationId: DeploymentService_GetDeploymentParameters summary: Get deployment parameters description: Returns the available gateways, runtime engines and versions, runtime config keys, and instance types for use when creating a deployment. parameters: - name: gatewayIds in: query description: "Optional set of gateway IDs to scope the returned parameters to. When provided,\ \ the response\n enumerates only the values valid across every listed gateway, so a deployment\ \ targeting this\n exact set and using only the returned values passes validation. Empty means\ \ all of the\n organization's gateways (the values are enumerated across them without the cross-gateway\n\ \ intersection)." schema: type: array items: type: string responses: '200': description: OK content: application/json: schema: $ref: '#/components/schemas/GetDeploymentParametersResponse' default: description: Default error response content: application/json: schema: $ref: '#/components/schemas/Status' tags: - DeploymentService /v1alpha1/inference/deployments/{id}: get: operationId: DeploymentService_GetDeployment summary: Get deployment description: Retrieves a single deployment by ID. parameters: - name: id in: path description: The ID of the deployment to get required: true schema: type: string responses: '200': description: OK content: application/json: schema: $ref: '#/components/schemas/GetDeploymentResponse' default: description: Default error response content: application/json: schema: $ref: '#/components/schemas/Status' tags: - DeploymentService delete: operationId: DeploymentService_DeleteDeployment summary: Delete deployment description: Deletes a deployment by ID. Returns the deleted deployment. parameters: - name: id in: path description: The ID of the deployment to delete required: true schema: type: string responses: '200': description: OK content: application/json: schema: $ref: '#/components/schemas/DeleteDeploymentResponse' default: description: Default error response content: application/json: schema: $ref: '#/components/schemas/Status' tags: - DeploymentService patch: operationId: DeploymentService_UpdateDeployment summary: Update deployment description: Updates an existing deployment. The request body replaces the deployment's spec; this endpoint does not accept a field mask. parameters: - name: id in: path description: The ID of the deployment to update required: true schema: type: string requestBody: required: true content: application/json: schema: $ref: '#/components/schemas/UpdateDeploymentRequest' responses: '200': description: OK content: application/json: schema: $ref: '#/components/schemas/UpdateDeploymentResponse' default: description: Default error response content: application/json: schema: $ref: '#/components/schemas/Status' tags: - DeploymentService /v1alpha1/inference/gateways: get: operationId: GatewayService_ListGateways summary: List gateways description: Returns every inference gateway that the caller can read, optionally filtered by update timestamp. parameters: - name: updatedAfter in: query description: Filter by those updated after this timestamp schema: type: string format: date-time responses: '200': description: OK content: application/json: schema: $ref: '#/components/schemas/ListGatewaysResponse' default: description: Default error response content: application/json: schema: $ref: '#/components/schemas/Status' tags: - GatewayService post: operationId: GatewayService_CreateGateway summary: Create gateway description: Creates a new inference gateway. Each gateway must be configured with exactly one authentication type and exactly one routing strategy. requestBody: required: true content: application/json: schema: $ref: '#/components/schemas/CreateGatewayRequest' responses: '200': description: OK content: application/json: schema: $ref: '#/components/schemas/CreateGatewayResponse' default: description: Default error response content: application/json: schema: $ref: '#/components/schemas/Status' tags: - GatewayService /v1alpha1/inference/gateways/parameters: get: operationId: GatewayService_GetGatewayParameters summary: Get gateway parameters description: Returns the available Availability Zones for use when creating a gateway. responses: '200': description: OK content: application/json: schema: $ref: '#/components/schemas/GetGatewayParametersResponse' default: description: Default error response content: application/json: schema: $ref: '#/components/schemas/Status' tags: - GatewayService /v1alpha1/inference/gateways/{id}: get: operationId: GatewayService_GetGateway summary: Get gateway description: Retrieves a single gateway by its UUID. parameters: - name: id in: path description: The unique identifier of the reservation to get, UUID format required: true schema: type: string responses: '200': description: OK content: application/json: schema: $ref: '#/components/schemas/GetGatewayResponse' default: description: Default error response content: application/json: schema: $ref: '#/components/schemas/Status' tags: - GatewayService delete: operationId: GatewayService_DeleteGateway summary: Delete gateway description: Deletes a gateway by ID. Returns the deleted gateway. parameters: - name: id in: path description: The unique identifier of the gateway to delete, UUID format required: true schema: type: string responses: '200': description: OK content: application/json: schema: $ref: '#/components/schemas/DeleteGatewayResponse' default: description: Default error response content: application/json: schema: $ref: '#/components/schemas/Status' tags: - GatewayService patch: operationId: GatewayService_UpdateGateway summary: Update gateway description: Updates an existing gateway. The request body replaces the gateway's spec; this endpoint does not accept a field mask. parameters: - name: id in: path description: The unique identifier of the gateway to update, UUID format required: true schema: type: string requestBody: required: true content: application/json: schema: $ref: '#/components/schemas/UpdateGatewayRequest' responses: '200': description: OK content: application/json: schema: $ref: '#/components/schemas/UpdateGatewayResponse' default: description: Default error response content: application/json: schema: $ref: '#/components/schemas/Status' tags: - GatewayService /v1alpha1/inference/hotLoads: get: operationId: HotLoadService_ListHotLoads summary: List hot-loads description: Returns the caller's hot-loads, newest first, optionally filtered to a single deployment. parameters: - name: parentDeploymentId in: query description: The ID of a deployment to filter by. When omitted, every hot-load in the caller's organization is returned. schema: type: string - name: maxPageSize in: query description: "The maximum number of hot-loads to return. The service may return fewer than this\ \ value. If unspecified or 0, at\n most 50 hot-loads are returned. The maximum value is 1000;\ \ values above 1000 are coerced to 1000. A negative\n value is rejected with INVALID_ARGUMENT." schema: type: integer format: int32 - name: pageToken in: query description: "A page token received from a previous ListHotLoads call. Provide this to retrieve\ \ the subsequent page.\n Use the same parent_deployment_id filter, including leaving it unset\ \ if it was unset in the previous call.\n A token used with a different filter is rejected with\ \ INVALID_ARGUMENT.\n The token is opaque and must not be constructed by clients." schema: type: string responses: '200': description: OK content: application/json: schema: $ref: '#/components/schemas/ListHotLoadsResponse' default: description: Default error response content: application/json: schema: $ref: '#/components/schemas/Status' tags: - HotLoadService post: operationId: HotLoadService_CreateHotLoad summary: Create hot-load description: Triggers a checkpoint hot-load on a deployment. The operation runs asynchronously, so a successful response means the request was accepted, not that the weights are loaded. requestBody: required: true content: application/json: schema: $ref: '#/components/schemas/CreateHotLoadRequest' responses: '200': description: OK content: application/json: schema: $ref: '#/components/schemas/CreateHotLoadResponse' default: description: Default error response content: application/json: schema: $ref: '#/components/schemas/Status' tags: - HotLoadService /v1alpha1/inference/hotLoads/{hotLoadId}/replicas: get: operationId: HotLoadService_ListHotLoadReplicas summary: List hot-load replicas description: Returns the replicas of one hot-load, ordered by zone and name, in pages. parameters: - name: hotLoadId in: path description: The ID of the hot-load whose replicas to list required: true schema: type: string - name: maxPageSize in: query description: "The maximum number of replicas to return. The service may return fewer than this\ \ value. If unspecified or 0,\n at most 50 replicas are returned. The maximum value is 1000;\ \ values above 1000 are coerced to 1000. A\n negative value is rejected with INVALID_ARGUMENT." schema: type: integer format: int32 - name: pageToken in: query description: "A page token received from a previous ListHotLoadReplicas call. Provide this to\ \ retrieve the subsequent page.\n Use the same hot_load_id and organization as the previous\ \ call; otherwise, the token is rejected with\n INVALID_ARGUMENT. The token is opaque and must\ \ not be constructed by clients." schema: type: string responses: '200': description: OK content: application/json: schema: $ref: '#/components/schemas/ListHotLoadReplicasResponse' default: description: Default error response content: application/json: schema: $ref: '#/components/schemas/Status' tags: - HotLoadService /v1alpha1/inference/hotLoads/{id}: get: operationId: HotLoadService_GetHotLoad summary: Get hot-load description: Returns a hot-load by ID, including its resolved snapshot chain and aggregate state. parameters: - name: id in: path description: The ID of the hot-load required: true schema: type: string responses: '200': description: OK content: application/json: schema: $ref: '#/components/schemas/GetHotLoadResponse' default: description: Default error response content: application/json: schema: $ref: '#/components/schemas/Status' tags: - HotLoadService /v1alpha1/inference/hotLoads/{id}:cancel: post: operationId: HotLoadService_CancelHotLoad summary: Cancel hot-load description: Cancels a hot-load. This operation is not yet implemented and currently returns an `Unimplemented` error. parameters: - name: id in: path description: The ID of the hot-load to cancel required: true schema: type: string requestBody: required: true content: application/json: schema: $ref: '#/components/schemas/CancelHotLoadRequest' responses: '200': description: OK content: application/json: schema: $ref: '#/components/schemas/CancelHotLoadResponse' default: description: Default error response content: application/json: schema: $ref: '#/components/schemas/Status' tags: - HotLoadService components: schemas: BodyBasedRouting: description: Body based routing configuration type: object properties: apiType: description: Well-known API type for routing. `API_TYPE_OPENAI` extracts the model name from the `model` field in the JSON request body (OpenAI-compatible behavior). The zero value `API_TYPE_UNSPECIFIED` is never valid to send. type: string format: enum enum: - API_TYPE_OPENAI required: - apiType CancelHotLoadRequest: description: Request for CancelHotLoad type: object properties: id: description: The ID of the hot-load to cancel type: string required: - id CancelHotLoadResponse: description: Response for CancelHotLoad type: object properties: hotLoad: description: The canceled hot-load allOf: - $ref: '#/components/schemas/HotLoad' readOnly: true CapacityClaim: description: CapacityClaim object with specification and status fields type: object properties: spec: description: The specification of the CapacityClaim allOf: - $ref: '#/components/schemas/CapacityClaimSpec' readOnly: true status: description: The status of the CapacityClaim allOf: - $ref: '#/components/schemas/CapacityClaimStatus' readOnly: true CapacityClaimResources: description: Resources to reserve within a CapacityClaim type: object properties: instanceId: description: The instance type to reserve, by ID specifier (for example, `gb200-4x`). Identifiers are case-sensitive and must use the lowercase form. Must be a valid instance type within at least one of the specified `zones`. type: string instanceCount: description: The number of instances to reserve. type: integer format: uint32 capacityType: description: The capacity type for the CapacityClaim. `CAPACITY_TYPE_MANAGED` reserves instances from CoreWeave's managed capacity pool and is the only value accepted when creating or updating a claim. `CAPACITY_TYPE_CUSTOMER` is deprecated; it is not accepted on create or update, but can still appear when reading older claims. This field cannot be changed after the claim is created. type: string format: enum enum: - CAPACITY_TYPE_CUSTOMER - CAPACITY_TYPE_MANAGED zones: description: The Availability Zones the CapacityClaim may use resources from. Zone identifiers are case-sensitive and must use the uppercase canonical form (for example, `US-WEST-04A`). The order of `zones` implies an allocation preference; earlier entries are preferred. type: array items: type: string required: - instanceId - instanceCount - capacityType - zones CapacityClaimSpec: description: CapacityClaimSpec contains the specification for a CapacityClaim type: object properties: id: description: The unique identifier of the CapacityClaim, UUID format type: string readOnly: true name: description: The human readable name of the CapacityClaim type: string readOnly: true resources: description: The specification for the resources to reserve allOf: - $ref: '#/components/schemas/CapacityClaimResources' readOnly: true organizationId: description: The organization ID that owns the CapacityClaim type: string readOnly: true CapacityClaimStatus: description: CapacityClaimStatus contains the status fields for a CapacityClaim type: object properties: createdAt: description: The time at which the CapacityClaim was created. type: string format: date-time readOnly: true updatedAt: description: The time at which the CapacityClaim was last updated. type: string format: date-time readOnly: true status: description: The overall status of the CapacityClaim. `STATUS_UNSPECIFIED` means the status has not been set. See the [Inference API overview](https://docs.coreweave.com/products/inference/reference/api-overview#status-values) for the meaning of each value. type: string format: enum enum: - STATUS_UNSPECIFIED - STATUS_CREATING - STATUS_UPDATING - STATUS_DELETING - STATUS_ERROR - STATUS_FAILED - STATUS_READY readOnly: true conditions: description: List of conditions representing detailed status information type: array items: $ref: '#/components/schemas/Condition' readOnly: true allocatedInstances: description: The number of instances currently allocated and available. type: integer format: uint32 readOnly: true pendingInstances: description: The number of instances being provisioned to fulfill the capacity claim. type: integer format: uint32 readOnly: true Condition: description: Condition represents a detailed status condition for resources type: object properties: type: description: Type of condition type: string readOnly: true status: description: 'The status of the condition: `STATUS_TRUE` or `STATUS_FALSE`, or `STATUS_UNSPECIFIED` when the status has not been set.' type: string format: enum enum: - STATUS_UNSPECIFIED - STATUS_FALSE - STATUS_TRUE readOnly: true lastUpdateTime: description: Last time the condition was updated type: string format: date-time readOnly: true reason: description: Reason for the condition's last transition type: string readOnly: true message: description: Human-readable message indicating details about the condition type: string readOnly: true zone: description: Zone associated with the condition type: string readOnly: true CoreWeaveAuth: description: Use CoreWeave IAM authentication, no additional configuration type: object properties: {} CreateCapacityClaimRequest: description: Request to create a CapacityClaim type: object properties: id: description: The unique identifier of the CapacityClaim, UUID format type: string name: description: The human readable name of the CapacityClaim type: string resources: description: The specification for the resources to reserve allOf: - $ref: '#/components/schemas/CapacityClaimResources' required: - resources CreateCapacityClaimResponse: description: Response for CreateCapacityClaim type: object properties: capacityClaim: description: The created CapacityClaim allOf: - $ref: '#/components/schemas/CapacityClaim' readOnly: true CreateDeploymentRequest: description: Request for CreateDeployment type: object properties: id: description: The unique identifier of the deployment, UUID format type: string name: description: The name of the deployment type: string gatewayIds: description: The gateways to associate the deployment with type: array items: type: string runtime: description: Runtime selection and configuration allOf: - $ref: '#/components/schemas/DeploymentRuntime' resources: description: Resource configuration for the deployment allOf: - $ref: '#/components/schemas/DeploymentResources' model: description: The model configuration allOf: - $ref: '#/components/schemas/DeploymentModel' autoscaling: description: The autoscaling configuration allOf: - $ref: '#/components/schemas/DeploymentAutoscaling' traffic: description: The traffic configuration for the deployment allOf: - $ref: '#/components/schemas/DeploymentTraffic' disabled: description: Disable the deployment type: boolean hotLoad: description: "The hot-load configuration for the deployment. Set it to enable checkpoint hot-loading;\ \ it is immutable\n after creation." allOf: - $ref: '#/components/schemas/DeploymentHotLoad' required: - name - gatewayIds - runtime - resources - model - autoscaling - traffic CreateDeploymentResponse: description: Response for CreateDeployment type: object properties: deployment: description: The deployment that was created allOf: - $ref: '#/components/schemas/Deployment' readOnly: true CreateGatewayRequest: description: Request for CreateGateway type: object properties: id: description: The unique identifier of the gateway, UUID format type: string name: description: "The human readable name of the gateway. Capped at 38 characters so the derived\n\ \ production FQDN (..gw.cwinference.com) fits within the 64-character\n X.509\ \ Common Name limit." type: string zones: description: "The zones to make the gateway available in, limits where deployments associated\ \ with the gateway may exist, no\n zones means all may be used." type: array items: type: string coreWeaveAuth: description: CoreWeave IAM allOf: - $ref: '#/components/schemas/CoreWeaveAuth' weightsAndBiasesAuth: description: Weights & Biases allOf: - $ref: '#/components/schemas/WeightsAndBiasesAuth' endpointConfiguration: description: Additional endpoint configuration options allOf: - $ref: '#/components/schemas/EndpointConfiguration' bodyBasedRouting: description: Body based routing allOf: - $ref: '#/components/schemas/BodyBasedRouting' headerBasedRouting: description: Header based routing allOf: - $ref: '#/components/schemas/HeaderBasedRouting' pathBasedRouting: description: Path based routing allOf: - $ref: '#/components/schemas/PathBasedRouting' required: - name CreateGatewayResponse: description: Response for CreateGateway type: object properties: gateway: description: The gateway that was created allOf: - $ref: '#/components/schemas/Gateway' readOnly: true CreateHotLoadRequest: description: Request for CreateHotLoad type: object properties: deploymentId: description: The ID of the deployment to hot-load type: string hotLoadId: description: "The ID to use for the hot-load, which becomes its permanent identifier. Must be\ \ a UUID. If omitted, one is\n generated. Creating a second hot-load with an ID that already\ \ exists in the caller's organization fails with\n ALREADY_EXISTS; an ID that is unavailable\ \ for any other reason fails with INVALID_ARGUMENT." type: string identity: description: "The identity of the target snapshot within the deployment. Must be a single path\ \ segment.\n Multiple hot-load operations may target the same snapshot identity." type: string type: description: The type of the target snapshot. `SNAPSHOT_TYPE_FULL` carries a complete set of model weights and starts a new snapshot chain. `SNAPSHOT_TYPE_INCREMENTAL` is a delta against an earlier snapshot and requires `incrementalSnapshotMetadata`. This field is required; the zero value is rejected. type: string format: enum enum: - SNAPSHOT_TYPE_FULL - SNAPSHOT_TYPE_INCREMENTAL incrementalSnapshotMetadata: description: Delta metadata, required when type is SNAPSHOT_TYPE_INCREMENTAL allOf: - $ref: '#/components/schemas/IncrementalSnapshotMetadata' promptCachePolicy: description: How the prompt cache is handled once the swap completes. `PROMPT_CACHE_POLICY_RESET_ALL` prevents reuse of any entry computed before the swap. `PROMPT_CACHE_POLICY_PRESERVE` allows reuse, knowingly stale under the new weights. The policy applies to requests admitted after the transition, not to requests already in flight. Omit the field rather than sending the zero value; omitting it applies `PROMPT_CACHE_POLICY_PRESERVE`. type: string format: enum enum: - PROMPT_CACHE_POLICY_RESET_ALL - PROMPT_CACHE_POLICY_PRESERVE required: - deploymentId - identity - type CreateHotLoadResponse: description: Response for CreateHotLoad type: object properties: hotLoad: description: The created hot-load allOf: - $ref: '#/components/schemas/HotLoad' readOnly: true DeleteCapacityClaimResponse: description: Response for DeleteCapacityClaim type: object properties: capacityClaim: description: The deleted CapacityClaim allOf: - $ref: '#/components/schemas/CapacityClaim' readOnly: true DeleteDeploymentResponse: description: Response for DeleteDeployment type: object properties: deployment: description: The deployment that was deleted allOf: - $ref: '#/components/schemas/Deployment' readOnly: true DeleteGatewayResponse: description: Response for DeleteGateway type: object properties: gateway: description: The gateway that was deleted allOf: - $ref: '#/components/schemas/Gateway' readOnly: true Deployment: description: Deployment object with specification and status fields type: object properties: spec: description: The specification of the deployment allOf: - $ref: '#/components/schemas/DeploymentSpec' readOnly: true status: description: The status of the deployment allOf: - $ref: '#/components/schemas/DeploymentStatus' readOnly: true DeploymentAutoscaling: description: Autoscaling configuration for the deployment type: object properties: min: description: The minimum number of replicas. Must be at least `1`; scale-to-zero is not supported. type: integer format: uint32 max: description: The maximum number of replicas. Must be greater than or equal to `min`. Set equal to `min` to disable autoscaling. type: integer format: uint32 priority: description: Scaling priority relative to other deployments, from `0` to `1000`. Higher values receive scaling preference during resource contention. type: integer format: uint32 capacityClasses: description: The capacity classes the autoscaler may use for this deployment. Set `CAPACITY_CLASS_RESERVED` to schedule replicas onto capacity reserved by a CapacityClaim; set `CAPACITY_CLASS_ON_DEMAND` to use shared on-demand capacity. Omit the zero value `CAPACITY_CLASS_UNSPECIFIED` rather than sending it. type: array items: type: string format: enum enum: - CAPACITY_CLASS_RESERVED - CAPACITY_CLASS_ON_DEMAND concurrency: description: Target concurrent requests per replica. Lower values reduce latency; higher values increase throughput. Must be at least `1`. type: integer format: uint32 required: - min - max DeploymentHotLoad: description: "DeploymentHotLoad configures checkpoint hot-loading for the deployment. Set it at\ \ deployment creation to enable\n hot-loading, or omit it to disable hot-loading. This configuration\ \ cannot be changed after creation." type: object properties: bucket: description: The CoreWeave Object Storage bucket that holds checkpoint snapshots type: string pathPrefix: description: "The path prefix within the bucket that holds checkpoint snapshots. Append the\ \ snapshot identity to this\n prefix to locate the snapshot's weights and metadata.\n Example:\ \ with bucket \"my-model\" and path_prefix \"my-checkpoints\", the weights and metadata of\ \ the\n snapshot with identity \"weight_v001\" should live under \"s3://my-model/my-checkpoints/weight_v001/\"" type: string transitionMode: description: How requests are handled while a checkpoint swap is in progress on a replica. `TRANSITION_MODE_ASYNC` pauses in-flight requests and resumes them after the swap, and queues new requests until the swap completes. This field is required, and `TRANSITION_MODE_ASYNC` is the only accepted value. type: string format: enum enum: - TRANSITION_MODE_ASYNC required: - bucket - transitionMode DeploymentModel: description: Configuration of the model type: object properties: name: description: The model name used in inference requests (and returned by the gateway's `/models` endpoint). When multiple deployments on the same gateway share a name, traffic is split between them by `traffic.weight`. type: string bucket: description: The CoreWeave AI Object Storage (CAIOS) bucket the model is stored in. type: string path: description: The path within `bucket` to the model and its configuration files. type: string required: - name - bucket - path DeploymentResourceParameters: description: The available parameters for resource configuration type: object properties: instanceTypes: description: "Available instance types. May be empty — e.g. when a gateway_ids filter selects\ \ gateways that\n share no common instance type, or when the organization has no gateways." type: array items: type: string readOnly: true DeploymentResources: description: Resource configuration for the deployment type: object properties: instanceType: description: The instance type to use for the deployment. The available values are returned by `GET /v1alpha1/inference/deployments/parameters`. type: string gpuCount: description: Number of GPUs needed per instance. CPU and RAM are automatically assigned based on the chosen GPU count. type: integer format: uint32 required: - instanceType - gpuCount DeploymentRuntime: description: Runtime selection and configuration for deployments type: object properties: engine: description: The inference engine to use (for example, `vllm`). The available engines are returned by `GET /v1alpha1/inference/deployments/parameters`. type: string version: description: The version of the engine. Required. Must be one of the engine's available versions returned by the parameters endpoint. type: string engineConfig: description: Engine-specific configuration arguments. The allowed keys per engine are returned under `runtimeParameters.runtimeConfigOptions` by the parameters endpoint. type: object additionalProperties: type: string engineEnv: description: Map of engine-specific environment variables to inject into the model runtime container type: object additionalProperties: type: string required: - engine - version DeploymentRuntimeParameters: description: The available parameters for runtime configuration type: object properties: runtimeVersions: description: The available runtime versions per engine type: object additionalProperties: $ref: '#/components/schemas/DeploymentRuntimeParameters_RuntimeVersions' readOnly: true runtimeConfigOptions: description: The available runtime config options per engine type: object additionalProperties: $ref: '#/components/schemas/DeploymentRuntimeParameters_RuntimeConfigOptions' readOnly: true engineEnvOptions: description: The available engine environment variable options per engine type: object additionalProperties: $ref: '#/components/schemas/DeploymentRuntimeParameters_EngineEnvOptions' readOnly: true DeploymentRuntimeParameters_EngineEnvOptions: description: The available engine environment variable options type: object properties: allowedNames: description: The allowed engine environment variable names. type: array items: type: string readOnly: true DeploymentRuntimeParameters_RuntimeConfigOptions: description: The available runtime config options type: object properties: allowedKeys: description: The allowed engine config keys type: array items: type: string readOnly: true DeploymentRuntimeParameters_RuntimeVersions: description: List of runtime versions type: object properties: versions: description: The version strings type: array items: type: string readOnly: true DeploymentSpec: description: DeploymentSpec object with configuration fields type: object properties: id: description: The ID of the deployment type: string readOnly: true name: description: The name of the deployment type: string readOnly: true gatewayIds: description: The IDs of the gateways the deployment is associated with. A deployment may be attached to multiple gateways; traffic from each gateway is routed by that gateway's strategy. type: array items: type: string readOnly: true runtime: description: Runtime selection and configuration. The available engines, versions, and per-engine config keys are returned by `GET /v1alpha1/inference/deployments/parameters`. allOf: - $ref: '#/components/schemas/DeploymentRuntime' readOnly: true resources: description: Resource configuration for the deployment. CPU and RAM are automatically assigned based on `gpuCount` for the chosen `instanceType`. allOf: - $ref: '#/components/schemas/DeploymentResources' readOnly: true model: description: The model configuration. The model is loaded from a path in a CoreWeave AI Object Storage (CAIOS) bucket. allOf: - $ref: '#/components/schemas/DeploymentModel' readOnly: true autoscaling: description: Autoscaling configuration. CoreWeave manages scaling within the configured `min` and `max` bounds; scale-to-zero is not supported. allOf: - $ref: '#/components/schemas/DeploymentAutoscaling' readOnly: true traffic: description: Traffic configuration for the deployment. When multiple deployments share a model name on the same gateway, `traffic.weight` controls the share of traffic each deployment receives. allOf: - $ref: '#/components/schemas/DeploymentTraffic' readOnly: true organizationId: description: The organization ID that owns the deployment. type: string readOnly: true disabled: description: When `true`, the deployment does not receive traffic and is not scaled. type: boolean readOnly: true hotLoad: description: The hot-load configuration for the deployment allOf: - $ref: '#/components/schemas/DeploymentHotLoad' readOnly: true DeploymentStatus: description: DeploymentStatus object for status fields type: object properties: createdAt: description: The time at which the deployment was created. type: string format: date-time readOnly: true updatedAt: description: The time at which the deployment was last updated. type: string format: date-time readOnly: true status: description: The overall status of the deployment. `STATUS_UNSPECIFIED` means the status has not been set. See the [Inference API overview](https://docs.coreweave.com/products/inference/reference/api-overview#status-values) for the meaning of each value. type: string format: enum enum: - STATUS_UNSPECIFIED - STATUS_CREATING - STATUS_UPDATING - STATUS_DELETING - STATUS_ERROR - STATUS_FAILED - STATUS_READY readOnly: true conditions: description: List of conditions representing detailed status information type: array items: $ref: '#/components/schemas/Condition' readOnly: true DeploymentTraffic: description: Configuration for traffic control to the deployment type: object properties: weight: description: The relative weight of traffic routed to this deployment compared to other deployments with the same model name on the same gateway. Weights are normalized into percentages. type: integer format: uint32 EndpointConfiguration: description: Configuration options for the gateway endpoint, DNS, TLS, etc. type: object properties: additionalDns: description: Additional DNS configurations, these DNS need to be configured to point to the gateway endpoint manually type: array items: type: string edgeProxyMode: description: 'How client traffic reaches the gateway. `EDGE_PROXY_MODE_MANAGED` proxies traffic through CoreWeave''s managed edge network: DDoS mitigation and WAF, with TLS terminated at the edge and the origin address kept out of public DNS. `EDGE_PROXY_MODE_DIRECT` connects clients straight to the gateway''s load balancer, with no edge filtering and the load balancer address published in DNS. `EDGE_PROXY_MODE_UNSPECIFIED` (or omitting the field) applies the platform default, currently managed.' type: string format: enum enum: - EDGE_PROXY_MODE_UNSPECIFIED - EDGE_PROXY_MODE_MANAGED - EDGE_PROXY_MODE_DIRECT allowedSourceIpRanges: description: "Client source IP ranges permitted to reach the gateway, in CIDR notation (IPv4\ \ or IPv6), up to\n 50 entries. An empty list applies no source restriction. Use a full-length\ \ prefix for a single\n address (e.g. \"203.0.113.5/32\")." type: array items: type: string Gateway: description: Gateway object with specification and status fields type: object properties: spec: description: The specification of the gateway allOf: - $ref: '#/components/schemas/GatewaySpec' readOnly: true status: description: The status of the gateway allOf: - $ref: '#/components/schemas/GatewayStatus' readOnly: true GatewaySpec: description: GatewaySpec contains the specification for a Gateway type: object properties: id: description: The unique identifier of the gateway, UUID format type: string readOnly: true name: description: "The human readable name of the gateway. Combined with organization_id and the\n\ \ deployment base domain it forms the gateway's public FQDN\n (..gw.).\ \ See CreateGatewayRequest.name and\n UpdateGatewayRequest.name for the length cap applied\ \ on write." type: string readOnly: true zones: description: The Availability Zones the gateway is deployed in. The field accepts an array, but gateways are currently limited to one zone. The available zones are returned by `GET /v1alpha1/inference/gateways/parameters`. type: array items: type: string readOnly: true coreWeaveAuth: description: CoreWeave IAM authentication. Validates inference requests using the same CoreWeave API access tokens used to call the management API. allOf: - $ref: '#/components/schemas/CoreWeaveAuth' weightsAndBiasesAuth: description: Weights & Biases authentication. Validates inference requests using W&B SaaS credentials. W&B self-hosted is not currently supported. allOf: - $ref: '#/components/schemas/WeightsAndBiasesAuth' organizationId: description: The organization ID that owns the gateway type: string readOnly: true endpointConfiguration: description: Optional endpoint configuration. Use `additionalDns` to register additional DNS names that you point at the gateway endpoint manually. allOf: - $ref: '#/components/schemas/EndpointConfiguration' bodyBasedRouting: description: Body based routing allOf: - $ref: '#/components/schemas/BodyBasedRouting' headerBasedRouting: description: Header based routing allOf: - $ref: '#/components/schemas/HeaderBasedRouting' pathBasedRouting: description: Path based routing allOf: - $ref: '#/components/schemas/PathBasedRouting' GatewayStatus: description: GatewayStatus contains the status fields for a Gateway type: object properties: createdAt: description: The time at which the gateway was created. type: string format: date-time readOnly: true updatedAt: description: The time at which the gateway was last updated. type: string format: date-time readOnly: true status: description: The overall status of the gateway. `STATUS_UNSPECIFIED` means the status has not been set. See the [Inference API overview](https://docs.coreweave.com/products/inference/reference/api-overview#status-values) for the meaning of each value. Once `STATUS_READY`, the gateway exposes one or more endpoint URLs in `endpoints`. type: string format: enum enum: - STATUS_UNSPECIFIED - STATUS_CREATING - STATUS_UPDATING - STATUS_DELETING - STATUS_ERROR - STATUS_FAILED - STATUS_READY readOnly: true conditions: description: List of conditions representing detailed status information type: array items: $ref: '#/components/schemas/Condition' readOnly: true endpoints: description: The endpoint URLs for the gateway. The default endpoint follows the pattern `https://[GATEWAY-NAME].[ORG-ID].gw.cwinference.com`; additional entries appear when `endpointConfiguration.additionalDns` is set. type: array items: type: string readOnly: true GetCapacityClaimParametersResponse: description: Response for GetCapacityClaimParameters type: object properties: zoneInstanceTypes: description: Map keyed by Availability Zone (for example, `US-WEST-04A`) whose values list the `instanceId` strings that are valid for a CapacityClaim in that zone. type: object additionalProperties: $ref: '#/components/schemas/GetCapacityClaimParametersResponse_InstanceTypes' readOnly: true GetCapacityClaimParametersResponse_InstanceTypes: description: List of available instance types type: object properties: instanceIds: description: List of instance type IDs type: array items: type: string readOnly: true GetCapacityClaimResponse: description: Response for GetCapacityClaim type: object properties: capacityClaim: description: The CapacityClaim allOf: - $ref: '#/components/schemas/CapacityClaim' readOnly: true GetDeploymentParametersResponse: description: Response for GetDeploymentParameters type: object properties: gatewayIds: description: The gateway IDs in the caller's organization that can be referenced by a new deployment. type: array items: type: string runtimeParameters: description: The available runtime parameters for the deployments allOf: - $ref: '#/components/schemas/DeploymentRuntimeParameters' readOnly: true resourceParameters: description: The available resource parameters for the deployments allOf: - $ref: '#/components/schemas/DeploymentResourceParameters' readOnly: true required: - gatewayIds GetDeploymentResponse: description: Response for GetDeployment type: object properties: deployment: description: The deployment allOf: - $ref: '#/components/schemas/Deployment' readOnly: true GetGatewayParametersResponse: description: Response for GetGatewayParameters type: object properties: zones: description: The Availability Zone identifiers that are valid in a gateway's `zones` field. type: array items: type: string required: - zones GetGatewayResponse: description: Response for GetGateway type: object properties: gateway: description: The gateway allOf: - $ref: '#/components/schemas/Gateway' readOnly: true GetHotLoadResponse: description: Response for GetHotLoad type: object properties: hotLoad: description: The requested hot-load allOf: - $ref: '#/components/schemas/HotLoad' readOnly: true GoogleProtobufAny: description: Contains an arbitrary serialized message along with a @type that describes the type of the serialized message. type: object properties: '@type': description: The type of the serialized message. type: string additionalProperties: true HeaderBasedRouting: description: Header based routing configuration, uses a header value to route requests to the model by name type: object properties: headerName: description: The HTTP header to read the model name from on each inference request. Must be 1-100 characters. type: string required: - headerName HotLoad: description: HotLoad object with specification and status fields type: object properties: spec: description: The specification of the hot-load allOf: - $ref: '#/components/schemas/HotLoadSpec' readOnly: true status: description: The status of the hot-load allOf: - $ref: '#/components/schemas/HotLoadStatus' readOnly: true HotLoadReplicaStatus: description: The per-replica progress of a hot-load type: object properties: name: description: The name of the replica type: string readOnly: true currentSnapshotIdentity: description: The snapshot identity the replica currently serves type: string readOnly: true state: description: The state of this replica's checkpoint transition. `PENDING_UPDATE` means the replica serves a different snapshot identity and its transition has not started. `UPDATING` means the transition is in progress. `READY` means the replica serves the target snapshot identity. `ERROR` means the transition failed, and `errorReason` carries the cause. type: string format: enum enum: - REPLICA_STATE_UNSPECIFIED - PENDING_UPDATE - UPDATING - READY - ERROR readOnly: true errorReason: description: The reason for an error state type: string readOnly: true transitionedAt: description: The last time the replica state changed type: string format: date-time readOnly: true zone: description: The zone where the replica runs type: string readOnly: true startedAt: description: "The time the replica began transitioning to the target snapshot, set when it first\ \ enters UPDATING. Unset\n while the replica is still PENDING_UPDATE." type: string format: date-time readOnly: true finishedAt: description: "The time the replica reached a terminal state (READY or ERROR). Unset until then,\ \ and unset again if a retry\n moves the replica back out of a terminal state." type: string format: date-time readOnly: true updatedAt: description: The time the replica status was last updated, whether or not its state changed type: string format: date-time readOnly: true HotLoadSpec: description: HotLoadSpec describes one immutable checkpoint hot-load operation type: object properties: id: description: The ID of the hot-load type: string readOnly: true deploymentId: description: The ID of the deployment this hot-load targets type: string readOnly: true organizationId: description: The organization ID that owns the hot-load type: string readOnly: true identity: description: "The identity of the target snapshot within the deployment. Must be a single path\ \ segment.\n Multiple hot-load operations may target the same snapshot identity." type: string readOnly: true type: description: The type of the target snapshot. type: string format: enum enum: - SNAPSHOT_TYPE_UNSPECIFIED - SNAPSHOT_TYPE_FULL - SNAPSHOT_TYPE_INCREMENTAL readOnly: true incrementalSnapshotMetadata: description: Delta metadata, required when type is SNAPSHOT_TYPE_INCREMENTAL allOf: - $ref: '#/components/schemas/IncrementalSnapshotMetadata' readOnly: true promptCachePolicy: description: How the prompt cache was handled once the swap completed. type: string format: enum enum: - PROMPT_CACHE_POLICY_UNSPECIFIED - PROMPT_CACHE_POLICY_RESET_ALL - PROMPT_CACHE_POLICY_PRESERVE readOnly: true snapshotChain: description: The resolved snapshot chain, oldest first. The first element is always a full snapshot. type: array items: $ref: '#/components/schemas/SnapshotRef' readOnly: true HotLoadStatus: description: HotLoadStatus describes the observed state of a hot-load type: object properties: createdAt: description: The time the hot-load was created type: string format: date-time readOnly: true updatedAt: description: The time the hot-load status last changed type: string format: date-time readOnly: true finishedAt: description: The time the hot-load reached a terminal state type: string format: date-time readOnly: true state: description: The aggregate state of the hot-load. `HOT_LOAD_STATE_PENDING` means the operation was accepted but replica transitions have not started, and `HOT_LOAD_STATE_IN_PROGRESS` means they are underway. `HOT_LOAD_STATE_COMPLETED` means every replica serves the target snapshot identity. `HOT_LOAD_STATE_FAILED` and `HOT_LOAD_STATE_CANCELED` are terminal, and replicas may serve different snapshots until a later hot-load completes. type: string format: enum enum: - HOT_LOAD_STATE_UNSPECIFIED - HOT_LOAD_STATE_PENDING - HOT_LOAD_STATE_IN_PROGRESS - HOT_LOAD_STATE_COMPLETED - HOT_LOAD_STATE_FAILED - HOT_LOAD_STATE_CANCELED readOnly: true conditions: description: Detailed status conditions type: array items: $ref: '#/components/schemas/Condition' readOnly: true canceledAt: description: The time cancellation was requested; set only for canceled hot-loads type: string format: date-time readOnly: true IncrementalSnapshotMetadata: description: Metadata describing how an incremental snapshot relates to its predecessor type: object properties: previousSnapshotIdentity: description: The identity of the snapshot this delta was computed against type: string compressionFormat: description: The compression format of the delta. `COMPRESSION_FORMAT_ZSTD` is a Zstandard-compressed XOR delta. This field is required, and `COMPRESSION_FORMAT_ZSTD` is the only accepted value. type: string format: enum enum: - COMPRESSION_FORMAT_ZSTD checksumFormat: description: The checksum format used to verify snapshot integrity. This field is required, and `CHECKSUM_FORMAT_ADLER32` is the only accepted value. type: string format: enum enum: - CHECKSUM_FORMAT_ADLER32 required: - previousSnapshotIdentity - compressionFormat - checksumFormat ListCapacityClaimsResponse: description: Response for ListCapacityClaims type: object properties: capacityClaims: description: The list of CapacityClaims type: array items: $ref: '#/components/schemas/CapacityClaim' readOnly: true ListDeploymentsResponse: description: Response for ListDeployments type: object properties: items: description: The list of deployments type: array items: $ref: '#/components/schemas/Deployment' readOnly: true ListGatewaysResponse: description: Response for ListGateways type: object properties: items: description: The list of gateways type: array items: $ref: '#/components/schemas/Gateway' readOnly: true ListHotLoadReplicasResponse: description: Response for ListHotLoadReplicas type: object properties: replicas: description: The replicas of the hot-load, ordered by (zone, name) type: array items: $ref: '#/components/schemas/HotLoadReplicaStatus' readOnly: true nextPageToken: description: A token that can be sent as page_token to retrieve the next page. Empty when the final page has been reached. type: string readOnly: true ListHotLoadsResponse: description: Response for ListHotLoads type: object properties: hotLoads: description: The hot-loads matching the request, newest first type: array items: $ref: '#/components/schemas/HotLoad' readOnly: true nextPageToken: description: A token that can be sent as page_token to retrieve the next page. Empty when the final page has been reached. type: string readOnly: true PathBasedRouting: description: Path-based routing configuration. The model name is read from the first segment of the URL path (for example, `/my-model/v1/chat/completions`). URL-encode model names that contain special characters. type: object properties: {} SnapshotRef: description: One element of a resolved snapshot chain type: object properties: identity: description: The identity of the snapshot type: string readOnly: true type: description: The type of this snapshot in the resolved chain. type: string format: enum enum: - SNAPSHOT_TYPE_UNSPECIFIED - SNAPSHOT_TYPE_FULL - SNAPSHOT_TYPE_INCREMENTAL readOnly: true compressionFormat: description: The compression format, set for incremental snapshots. type: string format: enum enum: - COMPRESSION_FORMAT_UNSPECIFIED - COMPRESSION_FORMAT_ZSTD readOnly: true checksumFormat: description: The checksum format, set for incremental snapshots. type: string format: enum enum: - CHECKSUM_FORMAT_UNSPECIFIED - CHECKSUM_FORMAT_ADLER32 readOnly: true hotLoadId: description: The ID of the HotLoad operation that introduced this snapshot into the chain. type: string readOnly: true Status: description: Standard error response. `code` is a [`google.rpc.Code`](https://cloud.google.com/apis/design/errors#error_codes); `message` is human-readable English; `details` carries machine-readable error details when present. type: object properties: code: description: The status code, which should be an enum value of [google.rpc.Code][google.rpc.Code]. type: integer format: int32 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 details: description: A list of messages that carry the error details. There is a common set of message types for APIs to use. type: array items: $ref: '#/components/schemas/GoogleProtobufAny' UpdateCapacityClaimRequest: description: Request to update a CapacityClaim type: object properties: id: description: The unique identifier of the CapacityClaim to update, UUID format type: string resources: description: The specification for the resources to reserve allOf: - $ref: '#/components/schemas/CapacityClaimResources' required: - id - resources UpdateCapacityClaimResponse: description: Response for UpdateCapacityClaim type: object properties: capacityClaim: description: The updated CapacityClaim allOf: - $ref: '#/components/schemas/CapacityClaim' readOnly: true UpdateDeploymentRequest: description: Request for UpdateDeployment type: object properties: id: description: The ID of the deployment to update type: string name: description: The name of the deployment type: string gatewayIds: description: The gateways to associate the deployment with type: array items: type: string runtime: description: Runtime selection and configuration allOf: - $ref: '#/components/schemas/DeploymentRuntime' resources: description: Resource configuration for the deployment allOf: - $ref: '#/components/schemas/DeploymentResources' model: description: The model configuration allOf: - $ref: '#/components/schemas/DeploymentModel' autoscaling: description: The autoscaling configuration allOf: - $ref: '#/components/schemas/DeploymentAutoscaling' traffic: description: The traffic configuration for the deployment allOf: - $ref: '#/components/schemas/DeploymentTraffic' disabled: description: Disable the deployment type: boolean hotLoad: description: 'The hot-load configuration for the deployment. Immutable: it must match the value set at creation.' allOf: - $ref: '#/components/schemas/DeploymentHotLoad' required: - id - name - gatewayIds - runtime - resources - model - autoscaling - traffic UpdateDeploymentResponse: description: Response for UpdateDeployment type: object properties: deployment: description: The updated deployment allOf: - $ref: '#/components/schemas/Deployment' readOnly: true UpdateGatewayRequest: description: Request for UpdateGateway type: object properties: id: description: The unique identifier of the gateway to update, UUID format type: string name: description: "The human readable name of the gateway. Capped at 38 characters, matching\n CreateGatewayRequest.name." type: string zones: description: "The zones to make the gateway available in, limits where deployments associated\ \ with the gateway may exist, no\n zones means all may be used." type: array items: type: string coreWeaveAuth: description: CoreWeave IAM allOf: - $ref: '#/components/schemas/CoreWeaveAuth' weightsAndBiasesAuth: description: Weights & Biases allOf: - $ref: '#/components/schemas/WeightsAndBiasesAuth' endpointConfiguration: description: Additional endpoint configuration options allOf: - $ref: '#/components/schemas/EndpointConfiguration' bodyBasedRouting: description: Body based routing allOf: - $ref: '#/components/schemas/BodyBasedRouting' headerBasedRouting: description: Header based routing allOf: - $ref: '#/components/schemas/HeaderBasedRouting' pathBasedRouting: description: Path based routing allOf: - $ref: '#/components/schemas/PathBasedRouting' required: - id - name UpdateGatewayResponse: description: Response for UpdateGateway type: object properties: gateway: description: The updated gateway allOf: - $ref: '#/components/schemas/Gateway' readOnly: true WeightsAndBiasesAuth: description: Use W&B auth type: object properties: apiKey: description: The W&B organization API key. Required if `serverUrl` is set. type: string serverUrl: description: Optional W&B server URL. Defaults to the W&B shared SaaS instance. type: string enableUsageReports: description: Send inference usage data to W&B for tracking. type: boolean enableRateLimiting: description: Enable W&B-controlled rate limiting for inference requests. type: boolean securitySchemes: bearerAuth: type: http scheme: bearer bearerFormat: JWT description: CoreWeave API access token sent as a bearer token. x-default: Bearer {API_ACCESS_TOKEN} tags: - name: CapacityClaimService description: Endpoints for creating, listing, getting, updating, and deleting CapacityClaim reservations of GPU hardware for inference deployments. - name: DeploymentService description: Endpoints for creating, listing, getting, updating, and deleting model deployments. Each deployment associates a model with one or more gateways and configures runtime, resources, autoscaling, and traffic. - name: GatewayService description: Endpoints for creating, listing, getting, updating, and deleting inference gateways. Gateways provide authentication, request routing, load balancing, and traffic splitting for one or more deployments. - name: HotLoadService description: Endpoints for creating, listing, getting, and canceling checkpoint hot-loads. A hot-load loads a new set of model weights into a running deployment's replicas without restarting the inference processes or recreating the replicas. security: - bearerAuth: [] ```