CH·02CLI reference
gpu operations cancel
Cancel a launch that has not started its instance yet
Cancel a launch that has not started its instance yet
Synopsis
Cancels an instance launch (the operation returned by gpu instances create or gpu deploy) while it is still pending or waiting on a supplier. Nothing is created or billed for a cancelled launch. If the supplier finishes creating the machine after the cancel, GPU.ai terminates it right away.
A launch that already created its instance cannot be cancelled. Delete the instance with gpu instances delete instead. Cancelling an operation that already finished changes nothing; the command prints the operation and exits non-zero, so a script can tell the cancel did not take effect.
gpu operations cancel <operation-id> [flags]
Examples
# Stop a launch that is still waiting on a supplier
gpu operations cancel 7a1e9d84-2c50-4f6b-9b31-0c5f2a6d8e19
Options
-h, --help help for cancel
Options inherited from parent commands
--api-base string API base URL (override with GPUAI_API_BASE env) (default "https://api.gpu.ai/v1")
--debug Enable debug logging to stderr
-o, --output string Output format: table|json (default table on TTY, json otherwise)
SEE ALSO
- gpu operations - Inspect async operations