Serverless model changes on 11 December 2026: Qwen3.8-Flash-Next-FP8 and Qwen3-Embedding-8B are being retired, and Mistral-Large-3-675B-Instruct-2512-NVFP4 is moving to dedicated models. See what to do.
Aliases:placements, placementCommands that act on one placement take its ID or name either as a positional argument or with --id. If you omit it, the CLI opens an interactive picker, so always pass it in scripts.The server commands take the placement the same way, and also open a picker for --server if you omit it. Placement, project, reservation, and server flags accept a name as well as an ID.
server reconcile — Rebuild a drifted server onto the placement’s image
CLI v7.0.0 moved the server commands under server: servers became server list, and server-reboot, server-stop, and server-reconcile became server reboot, server stop, and server reconcile. The old names still work but are hidden from help.
list
List placements, optionally filtered by organization, project, or reservation.Aliases:list, ls
nscale placements list [flags]
Flags
Flag
Description
Organization ID or name
Project ID or name
Reservation ID or name
Emit the full JSON payload (mutually exclusive with -q)
nscale placements get <placement-id> -q '.status.readyHostCount' -q '.spec.count'
create
Create a new placement. Accepts input from a JSON file or stdin, with an optional cloud-init configuration for the placed servers.An SSH certificate authority is optional. To bind one, set the payload’s spec.serverSpec.sshCertificateAuthorityId field or pass --ssh-ca-id, which overrides the payload. Use nscale ssh-cas list to find it. Passing --ssh-ca-id "" removes an SSH CA set in the payload. When you create a placement interactively and your organization has SSH CAs, the CLI asks whether to bind one.
nscale placements create [flags]
Flags
Flag
Description
Path to a JSON file
Read JSON from standard input
Path to a cloud-init YAML file
SSH certificate authority ID or name to bind
Preview actions and payload without persisting
Automatically confirm placement creation
Emit the full JSON payload (mutually exclusive with -q)
Change a placement’s server image or update strategy. The CLI reads the placement, changes only the image and strategy, and writes it back, keeping every other field. All other placement fields are fixed after creation. Scripts pass --image, --strategy, or both. In a terminal, the CLI asks for what you leave out, as described below.Changing the image marks the placement’s servers as drifted. How they move onto the new image depends on the update strategy:
manual (the default for a new placement): servers keep their current image until you rebuild each one with server reconcile.
rolling: the service rebuilds drifted servers for you, at most --max-unavailable at a time. --strategy rolling requires --max-unavailable, and --max-unavailable is only accepted with --strategy rolling.
If you omit --strategy, the placement keeps the strategy it already has, so an image change on a rolling placement starts a rollout. On a manual placement, an interactive run asks whether to roll the new image out instead, and how many servers to rebuild at once. With --yes, or outside a terminal, it keeps manual.--image takes an image ID or name from the placement’s region. When the reservation reports its hosts’ CPU architecture, the CLI refuses an image built for a different one. When it doesn’t, any image in the region is accepted, and the API doesn’t check either, so confirm the architecture yourself before a rebuild. Without --image or --strategy, an interactive run lets you pick from the images the hosts can boot.
Rebuilding a server onto a new image wipes its local disk. Under rolling, that happens to every server, --max-unavailable at a time, as soon as the update is accepted. Back up anything on local disk before you change the image.
nscale placements update <placement-id> [flags]
Flags
Flag
Description
Placement ID or name
Organization ID or name
Image ID or name to converge the servers onto, from the placement’s region and host architecture
Update strategy: manual or rolling. The API’s spellings, Manual and RollingUpdate, also work.
Servers rebuilt at once under : a count (1) or a percentage of the placement’s servers (25%, rounded up)
Preview actions and payload without persisting
Automatically confirm placement update
Emit the full JSON payload (mutually exclusive with -q)
# Roll a new image out, one server at a timenscale placements update <placement-id> --image <image> --strategy rolling --max-unavailable 1 --yes# Change the image but rebuild servers yourself with server reconcilenscale placements update <placement-id> --image <image> --strategy manual --yes# Preview the request body without sending itnscale placements update <placement-id> --image <image> --dry-run
nscale placements server list <placement-id> --org <org-id>
Aliases:list, lsThe table shows each server’s name, private IP, GPUs, CPUs, and status, followed by a status summary line such as Status summary: rebuilding: 1, running: 3. The summary appears only in the human table; --json, -q, and the agent format are unchanged.The status is one of creating, ready, running, starting, stopping, stopped, queued, building, rebuilding, degraded, deleting, error, or unknown. rebuilding means the server is being rebuilt onto a new image, for example by server reconcile or a rolling update. Any other power state reported by the service is shown in lowercase.
Flags
Flag
Description
Placement ID or name
Organization ID or name
Emit the full JSON payload (mutually exclusive with -q)
Show at most N results (client-side; every result is still fetched). Cuts before -q. 0 for all
Examples
nscale placements server list <placement-id> -q '.[].metadata.name' -q '.[].status.privateIP'
After an image change under the manual strategy, list the servers that haven’t moved onto the new image yet:
nscale placements server list <placement-id> \ -q '.[] | select(.status.driftState == "Drifted") | .metadata.id'
server reboot
Reboot an existing server in a placement. Use --hard for a hard reboot.
nscale placements server reboot <placement-id> --server <server-id> --org <org-id> [flags]
Flags
Flag
Description
--id string
Placement ID or name
--server string
Server ID or name
--org string
Organization ID or name
--hard
Use hard reboot
--dry-run
Preview actions and payload without persisting
-y, --yes
Automatically confirm server reboot
Examples
nscale placements server reboot <placement-id> --server <server-id> --org <org-id> --yesnscale placements server reboot <placement-id> --server <server-id> --hard --yes
server stop
Stop an existing server in a placement.
nscale placements server stop <placement-id> --server <server-id> --org <org-id> [flags]
Flags
Flag
Description
--id string
Placement ID or name
--server string
Server ID or name
--org string
Organization ID or name
--dry-run
Preview actions without persisting
-y, --yes
Automatically confirm server stop
Example
nscale placements server stop <placement-id> --server <server-id> --org <org-id> --yes
server reconcile
Rebuild one drifted server onto the placement’s image. The server stays on the same host, but its local disk contents are lost. Use this after an image change under the manualupdate strategy to move servers onto the new image one at a time. If the server already runs the placement’s image, nothing changes.
nscale placements server reconcile <placement-id> --server <server-id> --org <org-id> [flags]
Flags
Flag
Description
--id string
Placement ID or name
--server string
Server ID or name
--org string
Organization ID or name
--dry-run
Preview actions without persisting
-y, --yes
Automatically confirm server reconcile
Example
nscale placements server reconcile <placement-id> --server <server-id> --org <org-id> --yes
Related
Reservations
Reserve GPU capacity and check available reservation units.
Was this page helpful?
Assistant
Responses are generated using AI and may contain mistakes.