---
title: "Deploy Rebalance"
method: POST
path: "/deploy/{deploy_id}/rebalance"
tags: ["Dedicated Models"]
---

# Deploy Rebalance

`POST /deploy/{deploy_id}/rebalance`

Start a GPU pool rebalance: move GPUs from this deployment onto another
deployment you own, one instance at a time and without downtime. Moving all
instances stops this deployment; start it again later to resume it.

## Path parameters

- `deploy_id` string, required

## Headers

- `xi-api-key` string, nullable
- `x-api-key` string, nullable

## Request body

- RebalanceIn
  - `target_deploy_id` string, required — Deployment to grow using GPUs freed from this one.
  - `count` integer, required — Number of target instances to create.
  - `dry_run` boolean — Validate and preview without moving anything.

## Response `200`

Successful Response

- RebalanceOut
  - `replace_id` string — Id of the started rebalance; empty on a dry run.
  - `gpu` string, required — GPU type being rebalanced.
  - `source_instances` integer, required — Source instances now.
  - `source_instances_after` integer, required — Source instances once the rebalance completes.
  - `target_instances` integer, required — Target instances now.
  - `target_instances_after` integer, required — Target instances once the rebalance completes.

## Other responses

- `400` — Bad Request
- `401` — Unauthorized
- `404` — Not Found
- `409` — Conflict
- `422` — Validation Error
- `503` — Service Unavailable

---

[API](https://skmtc.net/deepinfra/apis/deepinfra-api.md) · [All operations](https://skmtc.net/deepinfra/apis/deepinfra-api/llms.txt) · [OpenAPI document](https://skmtc-service-staging.skmtc.workers.dev/v1/apis/deepinfra/deepinfra-api/versions/2df28210837a/schema)