> ## 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. # List available runners > List the runners visible to the caller's organization, with optional view, capability, and visibility filters. ## OpenAPI ````yaml /openapi/sandboxes/openapi.yaml get /v1/availableRunners openapi: 3.0.3 info: title: CoreWeave Sandboxes API version: 0.0.1 description: >- The Sandboxes API lets you manage runners, templates, volumes, network configurations, organization settings, and the lifecycle of individual sandboxes on CoreWeave. servers: - url: https://api.cwsandbox.com description: >- CoreWeave sandbox data plane. Hosts the sandbox lifecycle, volumes, snapshots, runner discovery, and organization settings. - url: https://api.coreweave.com description: >- CoreWeave control plane. Hosts managed runner and sandbox template administration. security: - bearerAuth: [] tags: - name: NetworkConfigService description: >- Create, get, list, update, and delete desired network configurations. Updating a configuration bound to a running sandbox changes its network rules without a restart. Requires API enablement and a runner with live network update support. - name: SandboxService description: >- Endpoints for the sandbox lifecycle: create (directly, from a template, or from a Compose file), get, list, exec, read and write files, snapshot, connect, purge, and delete. - name: DiscoveryService description: >- Read-only endpoints for discovering the runners visible to the caller's organization, with view, capability, and service-visibility filters. Runner IDs are unique only within an organization. - name: SettingsService description: >- Organization-level sandbox settings: the object-storage Workload Identity Federation (WIF) configuration, the file-system snapshot bucket configuration, Compose build storage, and secret stores. - name: VolumeService description: >- CRUD operations for organization-registered volumes that sandboxes can mount, plus the validation call that confirms a volume's backing source is usable. - name: RunnerManagementService description: >- CRUD operations for managed runners and the runner policy each one carries. Policy is a field on the runner, mutated through `CreateManagedRunner` and `UpdateManagedRunner`. - name: SandboxTemplateService description: >- CRUD operations for organization-scoped sandbox templates, the partial specs that `CreateSandboxFromTemplate` merges with per-request overrides. paths: /v1/availableRunners: get: tags: - DiscoveryService summary: List available runners description: >- ListAvailableRunners lists the runners visible to the caller's organization, optionally filtered. RunnerView defaults to BASIC when unspecified. operationId: DiscoveryService_ListAvailableRunners parameters: - name: runnerGroupId in: query description: Filters (all optional; AND-combined). schema: type: string - name: gpuType in: query schema: type: string - name: architecture in: query schema: type: string - name: healthyOnly in: query schema: type: boolean - name: pageSize in: query schema: type: integer format: int32 - name: pageToken in: query schema: type: string - name: view in: query description: Detail level. Unspecified defaults to BASIC for List. schema: type: string format: enum enum: - RUNNER_VIEW_UNSPECIFIED - RUNNER_VIEW_BASIC - RUNNER_VIEW_FULL - name: serviceVisibility in: query description: >- Require a runner that can enforce this generic L4 service visibility. UNSPECIFIED does not filter. schema: enum: - VISIBILITY_UNSPECIFIED - VISIBILITY_PUBLIC - VISIBILITY_PRIVATE - VISIBILITY_CUSTOM type: string format: enum responses: '200': description: OK content: application/json: schema: $ref: >- #/components/schemas/coreweave.sandbox.v1.ListAvailableRunnersResponse default: description: Default error response content: application/json: schema: $ref: '#/components/schemas/google.rpc.Status' servers: - url: https://api.cwsandbox.com description: CoreWeave sandbox data plane. components: schemas: coreweave.sandbox.v1.ListAvailableRunnersResponse: type: object properties: runners: type: array items: $ref: '#/components/schemas/coreweave.sandbox.v1.AvailableRunner' nextPageToken: type: string google.rpc.Status: 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). 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/google.protobuf.Any' coreweave.sandbox.v1.AvailableRunner: description: |- AvailableRunner is a caller-facing view of a runner. Sensitive internals (cluster, namespace) are omitted. type: object properties: runnerId: type: string readOnly: true runnerGroupId: type: string readOnly: true tags: type: array items: type: string readOnly: true healthy: type: boolean readOnly: true capabilities: description: >- Reported features and aggregate capacity ceilings; included at BASIC and FULL views when capability information is available. allOf: - $ref: >- #/components/schemas/coreweave.sandbox.v1.RunnerCapabilitySummary readOnly: true resources: description: >- Dynamic availability; populated at FULL view for owned runners only (unset for shared-org runners). allOf: - $ref: '#/components/schemas/coreweave.sandbox.v1.RunnerResourceSummary' readOnly: true connectedAt: type: string format: date-time readOnly: true isShared: description: >- True if the runner belongs to a shared organization (resources omitted). type: boolean readOnly: true organizationId: description: |- Organization that owns the runner. Runner identity is the pair (organization_id, runner_id); runner_id alone is not globally unique. type: string readOnly: true google.protobuf.Any: 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 coreweave.sandbox.v1.RunnerCapabilitySummary: description: >- RunnerCapabilitySummary describes a runner's reported features and aggregate capacity ceilings, rather than its currently available resources. The max_cpu_millicores, max_memory_bytes, and max_gpu_count values can change between responses. Depending on runner configuration, CPU and memory ceilings may exclude capacity reserved for GPU workloads and can change without nodes being added or removed. These values do not guarantee that a placement fits. When capability information is available, feature fields are exposed for both owned and shared runners. The max_* values are disclosed only for owned runners; they remain zero for shared runners (AvailableRunner.is_shared). A zero value alone does not distinguish zero capacity from an unreported or undisclosed ceiling. When capabilities and resources are both present, use values from the same response rather than combining resources with cached ceilings. They need not reflect the same measurement time. type: object properties: maxCpuMillicores: description: Reported aggregate CPU capacity ceiling, in millicores. type: string readOnly: true maxMemoryBytes: description: Reported aggregate memory capacity ceiling, in bytes. type: string readOnly: true maxGpuCount: description: Reported aggregate GPU capacity, including GPUs currently in use. type: string readOnly: true supportedGpuTypes: type: array items: type: string readOnly: true supportedArchitectures: type: array items: type: string readOnly: true supportsPrivileged: type: boolean readOnly: true availableStorageClasses: type: array items: type: string readOnly: true supportedServiceVisibilities: description: >- Generic L4 service visibilities this runner can enforce through the v1 Service path. VISIBILITY_UNSPECIFIED denotes ports that are not reachable from outside the sandbox. Operator-defined beta mode names are never exposed. type: array items: enum: - VISIBILITY_UNSPECIFIED - VISIBILITY_PUBLIC - VISIBILITY_PRIVATE - VISIBILITY_CUSTOM type: string format: enum readOnly: true coreweave.sandbox.v1.RunnerResourceSummary: description: >- RunnerResourceSummary is a runner's dynamic, currently-available resources. available_cpu_millicores and available_memory_bytes are usually the measured availability, but for a short window after the runner's capacity changes they may instead be reported at the runner's advertised ceiling while that ceiling is being republished. In that window a value derived as ceiling minus available can read zero even though the runner has usage. available_gpu_count is always the measurement. When comparing with capacity ceilings, use capabilities from the same response when present. The values need not reflect the same measurement time. type: object properties: availableCpuMillicores: description: >- Measured available CPU in millicores, or the advertised CPU ceiling during the window described above. type: string readOnly: true availableMemoryBytes: description: >- Measured available memory in bytes, or the advertised memory ceiling during the window described above. type: string readOnly: true availableGpuCount: description: Measured available GPUs; never adjusted to a ceiling. type: string readOnly: true runningSandboxes: type: integer format: int32 readOnly: true securitySchemes: bearerAuth: type: http scheme: bearer bearerFormat: JWT description: CoreWeave API access token sent as a bearer token. x-default: Bearer {API_ACCESS_TOKEN} ````