> ## Documentation Index > Fetch the complete documentation index at: https://docs.sfcompute.com/llms.txt > Use this file to discover all available pages before exploring further. # Update multiple instances > > ⚠️ This endpoint is in [public preview](/preview/roadmap). Update one or more instances atomically. All listed instances must be in the same workspace; mixed-workspace batches are rejected with 422. ## OpenAPI ````yaml /preview/openapi.json patch /preview/v2/instances openapi: 3.1.0 info: title: sfc-api description: >- Public preview API - subject to change. See https://docs.sfcompute.com/preview/roadmap for details. version: 0.1.0 servers: - url: https://api.sfcompute.com security: - bearer_auth: [] tags: - name: Pools description: A bucket of owned compute balance over time. - name: Orders description: >- Place orders targeting a capacity to increase your reserved compute balance during some time period. - name: Instance Templates description: Reusable instance configuration. - name: Images description: Custom machine images for instances. - name: Roles description: TOML-based permission role definitions. - name: Grants description: Bind principals (users or tokens) to roles on a workspace. - name: Tokens description: Workspace-scoped API tokens. - name: Instances description: Spin up instances in a capacity to use your available compute. - name: Instance SKU Catalog description: Browse available instance SKU property definitions. - name: Procurements description: Market automations that maintain capacity by placing buy/sell orders. - name: Deployments description: Deployment automations that maintain a fleet of instances on a capacity. - name: Users description: Read-only access to users within the caller's organization. - name: Workspaces description: Resource containers scoped to an account. - name: Permissions description: Inspect what the caller is allowed to do. - name: Billing description: Billing profile, contacts, and auto top-up settings. - name: Orderbook description: >- Read-only orderbook visibility: bid/ask spread, depth of book, and historical fills, keyed on hardware requirements + delivery window. - name: Orders description: >- Estimate an order before placing it: filled price, fee, and operational notices. paths: /preview/v2/instances: patch: tags: - Instances summary: Update multiple instances description: >- > ⚠️ This endpoint is in [public preview](/preview/roadmap). Update one or more instances atomically. All listed instances must be in the same workspace; mixed-workspace batches are rejected with 422. operationId: batch_patch_instances requestBody: content: application/json: schema: $ref: '#/components/schemas/BatchPatchInstancesRequest' required: true responses: '200': description: Instances updated. content: application/json: schema: $ref: '#/components/schemas/BatchPatchInstancesResponse' '401': description: Unauthorized. content: application/json: schema: $ref: '#/components/schemas/UnauthorizedError' '403': description: Forbidden. content: application/json: schema: $ref: '#/components/schemas/ForbiddenError' '422': description: >- Unknown instance, cross-workspace batch, or batch over the entry cap. content: application/json: schema: $ref: '#/components/schemas/UnprocessableEntityError' '500': description: Internal server error. content: application/json: schema: $ref: '#/components/schemas/InternalServerError' components: schemas: BatchPatchInstancesRequest: type: object description: >- Request body for `PATCH /v2/instances` (batch). Each entry in `data` applies a partial patch to one instance; instances not mentioned are untouched. All entries must succeed or none — a single failure rolls back every other entry's writes (422). When the same `id` appears more than once in `data`, the last occurrence wins. required: - data properties: data: type: array items: $ref: '#/components/schemas/BatchPatchInstanceEntry' BatchPatchInstancesResponse: type: object description: >- Response shape for `PATCH /v2/instances` (batch). Mirrors the input list — one `InstanceResponse` per *unique* `id` in the request body, reflecting the post-write state. Unlike the paginated list response, there's no `cursor` or `has_more`: the response is exactly the instances the caller mentioned, no pagination involved. required: - object - data properties: object: type: string const: list default: list readOnly: true data: type: array items: $ref: '#/components/schemas/InstanceResponse' UnauthorizedError: type: object required: - error properties: error: type: object required: - type - message properties: type: type: string const: authentication_error default: authentication_error readOnly: true message: type: string x-speakeasy-error-message: true ForbiddenError: type: object required: - error properties: error: type: object required: - type - message properties: type: type: string const: forbidden default: forbidden readOnly: true message: type: string x-speakeasy-error-message: true UnprocessableEntityError: type: object required: - error properties: error: type: object required: - type - message properties: type: type: string const: unprocessable_entity default: unprocessable_entity readOnly: true message: type: string x-speakeasy-error-message: true details: type: array items: $ref: '#/components/schemas/ErrorDetail' InternalServerError: type: object required: - error properties: error: type: object required: - type - message properties: type: type: string const: api_error default: api_error readOnly: true message: type: string x-speakeasy-error-message: true BatchPatchInstanceEntry: type: object description: >- One entry in a [`BatchPatchInstancesRequest`]: the instance to patch (`id`) and the fields to update. `priority_level` is the only field currently patchable here. To update name, tags, or other instance fields, use `PATCH /v2/instances/{id}`. required: - id - priority_level properties: id: $ref: '#/components/schemas/InstanceId' priority_level: $ref: '#/components/schemas/InstancePriority' description: Instance priority arm to apply to this instance. Required. InstanceResponse: allOf: - $ref: '#/components/schemas/InstanceScope' - type: object required: - object - status - capacity - pool - created_at - image - cloud_init_user_data_used properties: object: type: string const: instance default: instance readOnly: true status: $ref: '#/components/schemas/InstanceStatus' instance_sku: oneOf: - type: 'null' - $ref: '#/components/schemas/InstanceSkuSummary' description: >- Instance SKU this instance is running on. Only present when assigned to a physical machine. Carries the SKU's human-readable name when one is registered. capacity: $ref: '#/components/schemas/CapacitySummary' description: Deprecated — use `pool`. The pool this instance is utilizing. pool: $ref: '#/components/schemas/PoolSummary' created_at: $ref: '#/components/schemas/UnixEpoch' image: $ref: '#/components/schemas/ImageSummary' description: Image this instance was launched from. deployment: oneOf: - type: 'null' - $ref: '#/components/schemas/DeploymentSummary' description: Deployment managing this instance, if any. cloud_init_user_data_used: type: boolean description: Whether cloud-init user data is configured for this instance. cloud_init_user_data: type: string format: byte description: >- Base64-encoded [cloud-init user data](https://cloudinit.readthedocs.io/en/latest/explanation/format/index.html). example: IyEvYmluL2Jhc2gKZWNobyBoZWxsbyB3b3JsZAo= contentEncoding: base64 tags: oneOf: - type: 'null' - $ref: '#/components/schemas/Tags' description: Metadata tags attached to this instance. expected_shutdown_at: oneOf: - type: 'null' - $ref: '#/components/schemas/UnixEpoch' description: >- Predicted Unix timestamp at which this node will be terminated because its capacity drops below the count of running nodes. `null` if no shutdown is scheduled in the predictable horizon — either the capacity covers this node indefinitely, or a future capacity increase blocks the prediction. Recomputed on every read. priority_level: oneOf: - type: 'null' - $ref: '#/components/schemas/InstancePriority' description: >- Instance priority. Lower-priority instances are terminated first when the capacity's quota drops below the running-instance count. ErrorDetail: type: object required: - code - message properties: field: type: - string - 'null' description: The field that caused the error (for validation errors) code: type: string description: Specific error code for this detail message: type: string description: Human-readable error message InstanceId: oneOf: - type: string examples: - inst_k3R-nX9vLm7Qp2Yw5Jd8F pattern: inst_[0-9a-zA-Z_-]{1,21} - type: string description: Legacy alias prefix; accepted on read, never emitted on write. pattern: vm_[0-9a-zA-Z_-]{1,21} description: >- Accepts the canonical prefix below; additional legacy prefixes are aliased for read compatibility. Writes always emit the canonical form. InstancePriority: type: string description: >- Instance priority — a relative ranking that determines which instances the system prefers to keep running when capacity is constrained. When a capacity's quota drops below its running-instance count, instances are terminated in priority order (lower first). Ordering: `yield < normal < preferred < critical`. enum: - yield - normal - preferred - critical InstanceScope: type: object required: - id - resource_path - owner - workspace - workspace_id - name properties: id: $ref: '#/components/schemas/InstanceId' resource_path: $ref: '#/components/schemas/instanceResourcePath_InstanceId' owner: $ref: '#/components/schemas/Name' workspace: $ref: '#/components/schemas/Name' workspace_id: $ref: '#/components/schemas/WorkspaceId' name: $ref: '#/components/schemas/Name' InstanceStatus: type: string description: >- `awaiting_allocation` when waiting for compute allocation on its capacity, `running` once assigned and the physical machine is running (still takes time for the image to be downloaded and booted), `terminated` when stopped by the user or after running out of allocation, `failed` on hardware fault. enum: - awaiting_allocation - running - terminated - failed InstanceSkuSummary: type: object description: >- A summary of an instance SKU - its `id` and human-recognizable `alias` - embedded on resources that reference a SKU. required: - object - id - alias properties: object: type: string const: instance_sku default: instance_sku readOnly: true id: $ref: '#/components/schemas/InstanceSkuId' alias: $ref: '#/components/schemas/Name' CapacitySummary: type: object required: - id - name properties: id: $ref: '#/components/schemas/CapacityId' name: $ref: '#/components/schemas/Name' PoolSummary: type: object description: A pool referenced by id and name. required: - id - name properties: id: $ref: '#/components/schemas/PoolId' name: $ref: '#/components/schemas/Name' UnixEpoch: type: integer format: int64 description: Unix timestamp. example: 1738972800 ImageSummary: type: object required: - id - name properties: id: $ref: '#/components/schemas/ImageId' name: $ref: '#/components/schemas/Name' DeploymentSummary: type: object required: - id - name properties: id: $ref: '#/components/schemas/DeploymentId' name: $ref: '#/components/schemas/Name' Tags: type: object additionalProperties: type: string examples: - prod maxLength: 256 pattern: ^[^,=]{0,256}$ propertyNames: type: string examples: - env maxLength: 128 minLength: 1 pattern: ^[^_,= ][^,= ]{0,127}$ examples: - env: prod team: infra maxProperties: 50 instanceResourcePath_InstanceId: type: string description: >- A resource path for a instance resource. Format: sfc:instance:::. examples: - sfc:instance::: pattern: sfc:instance:([a-zA-Z0-9._-]+:){2}[a-zA-Z0-9._-]+ Name: type: string examples: - my-resource-name maxLength: 255 minLength: 1 pattern: '[a-zA-Z0-9][a-zA-Z0-9._-]{0,254}' WorkspaceId: type: string examples: - wksp_k3R-nX9vLm7Qp2Yw5Jd8F pattern: wksp_[0-9a-zA-Z_-]{1,21} InstanceSkuId: oneOf: - type: string examples: - isku_k3R-nX9vLm7Qp2Yw5Jd8F pattern: isku_[0-9a-zA-Z_-]{1,21} - type: string description: Legacy alias prefix; accepted on read, never emitted on write. pattern: clus_[0-9a-zA-Z_-]{1,21} description: >- Accepts the canonical prefix below; additional legacy prefixes are aliased for read compatibility. Writes always emit the canonical form. CapacityId: oneOf: - type: string examples: - cap_k3R-nX9vLm7Qp2Yw5Jd8F pattern: cap_[0-9a-zA-Z_-]{1,21} - type: string description: Legacy alias prefix; accepted on read, never emitted on write. pattern: pool_[0-9a-zA-Z_-]{1,21} description: >- Accepts the canonical prefix below; additional legacy prefixes are aliased for read compatibility. Writes always emit the canonical form. PoolId: type: string examples: - pool_k3R-nX9vLm7Qp2Yw5Jd8F pattern: pool_[0-9a-zA-Z_-]{1,21} ImageId: oneOf: - type: string examples: - image_k3R-nX9vLm7Qp2Yw5Jd8F pattern: image_[0-9a-zA-Z_-]{1,21} - type: string description: Legacy alias prefix; accepted on read, never emitted on write. pattern: vmi_[0-9a-zA-Z_-]{1,21} description: >- Accepts the canonical prefix below; additional legacy prefixes are aliased for read compatibility. Writes always emit the canonical form. DeploymentId: type: string examples: - depl_k3R-nX9vLm7Qp2Yw5Jd8F pattern: depl_[0-9a-zA-Z_-]{1,21} securitySchemes: bearer_auth: type: http scheme: bearer bearerFormat: JWT description: >- Create an API token using `sf tokens create` or at https://sfcompute.com/account/api-keys. ````