Scale or redeploy a model

Changes a model’s deployment incrementally. instances merges into the current plan: a listed region takes the new count, an omitted region is untouched, and count: 0 scales a region down.

release is optional. Omit it to change capacity only and keep the deployed release. Supply it to roll every region to that release, not only the regions in the plan. The body must set release, instances, or both.

strategy and immediate behave as on PUT. Without a release, REPLACE_ALL is treated as ROLLING, and immediate is refused.

Callable by an account_admin or developer of the owning account when model serving is enabled for the account. The CLI equivalent is reactor model deploy.

PATCH

models

{id}

releases

Scale or redeploy a model

curl --request PATCH \
  --url https://api.reactor.inc/models/{id}/releases \
  --header 'Authorization: Bearer <token>' \
  --header 'Content-Type: application/json' \
  --data '
{
  "instances": [
    {
      "region": "us-west",
      "count": 5
    }
  ]
}
'

import requests

url = "https://api.reactor.inc/models/{id}/releases"

payload = { "instances": [
        {
            "region": "us-west",
            "count": 5
        }
    ] }
headers = {
    "Authorization": "Bearer <token>",
    "Content-Type": "application/json"
}

response = requests.patch(url, json=payload, headers=headers)

print(response.text)

const options = {
  method: 'PATCH',
  headers: {Authorization: 'Bearer <token>', 'Content-Type': 'application/json'},
  body: JSON.stringify({instances: [{region: 'us-west', count: 5}]})
};

fetch('https://api.reactor.inc/models/{id}/releases', options)
  .then(res => res.json())
  .then(res => console.log(res))
  .catch(err => console.error(err));

<?php

$curl = curl_init();

curl_setopt_array($curl, [
  CURLOPT_URL => "https://api.reactor.inc/models/{id}/releases",
  CURLOPT_RETURNTRANSFER => true,
  CURLOPT_ENCODING => "",
  CURLOPT_MAXREDIRS => 10,
  CURLOPT_TIMEOUT => 30,
  CURLOPT_HTTP_VERSION => CURL_HTTP_VERSION_1_1,
  CURLOPT_CUSTOMREQUEST => "PATCH",
  CURLOPT_POSTFIELDS => json_encode([
    'instances' => [
        [
                'region' => 'us-west',
                'count' => 5
        ]
    ]
  ]),
  CURLOPT_HTTPHEADER => [
    "Authorization: Bearer <token>",
    "Content-Type: application/json"
  ],
]);

$response = curl_exec($curl);
$err = curl_error($curl);

curl_close($curl);

if ($err) {
  echo "cURL Error #:" . $err;
} else {
  echo $response;
}

package main

import (
	"fmt"
	"strings"
	"net/http"
	"io"
)

func main() {

	url := "https://api.reactor.inc/models/{id}/releases"

	payload := strings.NewReader("{\n  \"instances\": [\n    {\n      \"region\": \"us-west\",\n      \"count\": 5\n    }\n  ]\n}")

	req, _ := http.NewRequest("PATCH", url, payload)

	req.Header.Add("Authorization", "Bearer <token>")
	req.Header.Add("Content-Type", "application/json")

	res, _ := http.DefaultClient.Do(req)

	defer res.Body.Close()
	body, _ := io.ReadAll(res.Body)

	fmt.Println(string(body))

}

HttpResponse<String> response = Unirest.patch("https://api.reactor.inc/models/{id}/releases")
  .header("Authorization", "Bearer <token>")
  .header("Content-Type", "application/json")
  .body("{\n  \"instances\": [\n    {\n      \"region\": \"us-west\",\n      \"count\": 5\n    }\n  ]\n}")
  .asString();

require 'uri'
require 'net/http'

url = URI("https://api.reactor.inc/models/{id}/releases")

http = Net::HTTP.new(url.host, url.port)
http.use_ssl = true

request = Net::HTTP::Patch.new(url)
request["Authorization"] = 'Bearer <token>'
request["Content-Type"] = 'application/json'
request.body = "{\n  \"instances\": [\n    {\n      \"region\": \"us-west\",\n      \"count\": 5\n    }\n  ]\n}"

response = http.request(request)
puts response.read_body

{
  "release": "v1.2.3-gabcdef0",
  "immediate": false,
  "revision": {
    "number": 4
  }
}

{
  "error": "<string>",
  "message": "<string>",
  "resource": "<string>",
  "reason": "<string>"
}

{
  "error": "<string>",
  "message": "<string>",
  "resource": "<string>",
  "reason": "<string>"
}

{
  "error": "<string>",
  "message": "<string>",
  "resource": "<string>",
  "reason": "<string>"
}

{
  "error": "<string>",
  "message": "<string>",
  "resource": "<string>",
  "reason": "<string>"
}

{
  "error": "accelerator_quota_exceeded",
  "message": "deployment plan would commit 12 'NVIDIA_H100' accelerators in region 'us-east', but only 10 are allocated to this account. Lower the requested count or request a larger allocation"
}

{
  "error": "<string>",
  "message": "<string>",
  "resource": "<string>",
  "reason": "<string>"
}

{
  "error": "<string>",
  "message": "<string>",
  "resource": "<string>",
  "reason": "<string>"
}

Authorizations

Authorization

string

header

required

A Reactor API key (rk_...), sent as Authorization: Bearer rk_.... Keep it on a server.

Path Parameters

string<uuid>

required

The model's ID (not its project ID). Find it with GET /models?name=your-org/your-model.

Body

application/json

release

string

The release to deploy. Required on PUT; optional on PATCH.

Example:

"v1.2.3-gabcdef0"

instances

object[]

The instance plan per region. On PUT it replaces the plan; on PATCH it merges. Omit to leave capacity unchanged.

Show child attributes

strategy

enum<string>

default:ROLLING

How the old release stops serving. Matched exactly.

Available options:

ROLLING,

REPLACE_ALL

immediate

boolean

default:false

Close the model's live sessions once the deploy is sent. Requires a release and REPLACE_ALL.

Response

The change was accepted. A capacity-only change omits the weights fields.

release

string

required

The release now deployed.

immediate

boolean

required

Whether this deploy closes live sessions.

weights_carried_from

string

The release whose weights were reused because this release has none.

weights_missing

boolean

true when the release deployed with no weights available. Omitted when false.

revision

object

The production revision this deploy created. Omitted when nothing changed.

Show child attributes

⌘I

Scale or redeploy a model

curl --request PATCH \
  --url https://api.reactor.inc/models/{id}/releases \
  --header 'Authorization: Bearer <token>' \
  --header 'Content-Type: application/json' \
  --data '
{
  "instances": [
    {
      "region": "us-west",
      "count": 5
    }
  ]
}
'

import requests

url = "https://api.reactor.inc/models/{id}/releases"

payload = { "instances": [
        {
            "region": "us-west",
            "count": 5
        }
    ] }
headers = {
    "Authorization": "Bearer <token>",
    "Content-Type": "application/json"
}

response = requests.patch(url, json=payload, headers=headers)

print(response.text)

const options = {
  method: 'PATCH',
  headers: {Authorization: 'Bearer <token>', 'Content-Type': 'application/json'},
  body: JSON.stringify({instances: [{region: 'us-west', count: 5}]})
};

fetch('https://api.reactor.inc/models/{id}/releases', options)
  .then(res => res.json())
  .then(res => console.log(res))
  .catch(err => console.error(err));

<?php

$curl = curl_init();

curl_setopt_array($curl, [
  CURLOPT_URL => "https://api.reactor.inc/models/{id}/releases",
  CURLOPT_RETURNTRANSFER => true,
  CURLOPT_ENCODING => "",
  CURLOPT_MAXREDIRS => 10,
  CURLOPT_TIMEOUT => 30,
  CURLOPT_HTTP_VERSION => CURL_HTTP_VERSION_1_1,
  CURLOPT_CUSTOMREQUEST => "PATCH",
  CURLOPT_POSTFIELDS => json_encode([
    'instances' => [
        [
                'region' => 'us-west',
                'count' => 5
        ]
    ]
  ]),
  CURLOPT_HTTPHEADER => [
    "Authorization: Bearer <token>",
    "Content-Type: application/json"
  ],
]);

$response = curl_exec($curl);
$err = curl_error($curl);

curl_close($curl);

if ($err) {
  echo "cURL Error #:" . $err;
} else {
  echo $response;
}

package main

import (
	"fmt"
	"strings"
	"net/http"
	"io"
)

func main() {

	url := "https://api.reactor.inc/models/{id}/releases"

	payload := strings.NewReader("{\n  \"instances\": [\n    {\n      \"region\": \"us-west\",\n      \"count\": 5\n    }\n  ]\n}")

	req, _ := http.NewRequest("PATCH", url, payload)

	req.Header.Add("Authorization", "Bearer <token>")
	req.Header.Add("Content-Type", "application/json")

	res, _ := http.DefaultClient.Do(req)

	defer res.Body.Close()
	body, _ := io.ReadAll(res.Body)

	fmt.Println(string(body))

}

HttpResponse<String> response = Unirest.patch("https://api.reactor.inc/models/{id}/releases")
  .header("Authorization", "Bearer <token>")
  .header("Content-Type", "application/json")
  .body("{\n  \"instances\": [\n    {\n      \"region\": \"us-west\",\n      \"count\": 5\n    }\n  ]\n}")
  .asString();

require 'uri'
require 'net/http'

url = URI("https://api.reactor.inc/models/{id}/releases")

http = Net::HTTP.new(url.host, url.port)
http.use_ssl = true

request = Net::HTTP::Patch.new(url)
request["Authorization"] = 'Bearer <token>'
request["Content-Type"] = 'application/json'
request.body = "{\n  \"instances\": [\n    {\n      \"region\": \"us-west\",\n      \"count\": 5\n    }\n  ]\n}"

response = http.request(request)
puts response.read_body

{
  "release": "v1.2.3-gabcdef0",
  "immediate": false,
  "revision": {
    "number": 4
  }
}

{
  "error": "<string>",
  "message": "<string>",
  "resource": "<string>",
  "reason": "<string>"
}

{
  "error": "<string>",
  "message": "<string>",
  "resource": "<string>",
  "reason": "<string>"
}

{
  "error": "<string>",
  "message": "<string>",
  "resource": "<string>",
  "reason": "<string>"
}

{
  "error": "<string>",
  "message": "<string>",
  "resource": "<string>",
  "reason": "<string>"
}

{
  "error": "accelerator_quota_exceeded",
  "message": "deployment plan would commit 12 'NVIDIA_H100' accelerators in region 'us-east', but only 10 are allocated to this account. Lower the requested count or request a larger allocation"
}

{
  "error": "<string>",
  "message": "<string>",
  "resource": "<string>",
  "reason": "<string>"
}

{
  "error": "<string>",
  "message": "<string>",
  "resource": "<string>",
  "reason": "<string>"
}

Get started

SDK Reference

CLI Reference

Platform API Reference

Endpoints

Scale or redeploy a model

Authorizations

Path Parameters

Body

Response