Computing resources

Manage the compute resource lifecycle: list, create, update, delete, suspend, resume, or retry resources, and inspect runtime and image information. Use the action page to confirm the resource state and available controls.

Choose a task

Task	Description
Find accessible instances	Lists the compute instances in the current workspace that the caller has read access to.
Inspect instance configuration and state	Read the configuration, capacity, and lifecycle status of a compute instance.
Add a compute instance	Create a compute instance in the current workspace.
Change instance configuration	Change instance configuration; Serverless workers still activate on subsequent workload demand.
Check prerequisites for deletion	Check the active tasks and workflow bindings that need to be processed before deleting the compute instance.
Remove a compute instance	Delete the compute instance in the current workspace.
Suspend an active instance	Pauses a compute instance in the ACTIVE state, stops its workers, and transitions the instance to SUSPENDED.
Resume a suspended instance	Return a suspended instance to IDLE without immediately deploying workers.
Retry an instance in error	Retry an instance in ERROR and return it to IDLE without immediate worker redeployment.
Inspect Worker replicas	Query the runtime replica status of a compute instance by Worker type.
Inspect instance usage metrics	Read usage metrics for a compute instance.
Find Worker images	Lists the Worker images that can be selected when creating or updating a compute instance.
Repair legacy resource ownership	Register legacy resource ownership as a controlled repair requiring administrative rights.

Last updated on Sep 10, 2026