Create Autojob Workergroup

Creates a new serverless workergroup (autojob) for the authenticated user, optionally creating an endpoint if one doesn’t exist.

POST /api/v0/autojobs/

Create Autojob Workergroup

cURL

curl --request POST \
  --url https://console.vast.ai/api/v0/autojobs/ \
  --header 'Authorization: Bearer <token>' \
  --header 'Content-Type: application/json' \
  --data '
{
  "min_load": 0.1,
  "min_cold_load": 0,
  "target_util": 0.8,
  "cold_mult": 2,
  "cold_workers": 3,
  "max_workers": 10,
  "test_workers": 1,
  "gpu_ram": 16,
  "autoscaler_instance": "prod",
  "docker_login_user": "myuser",
  "docker_login_pass": "<string>",
  "docker_login_repo": "docker.io",
  "template_hash": "abc123def456",
  "template_id": 999,
  "launch_args": "--image pytorch/pytorch:latest --disk 20",
  "endpoint_name": "my-inference-endpoint",
  "endpoint_id": 42,
  "search_params": "gpu_ram>=16 num_gpus=1"
}
'

import requests

url = "https://console.vast.ai/api/v0/autojobs/"

payload = {
    "min_load": 0.1,
    "min_cold_load": 0,
    "target_util": 0.8,
    "cold_mult": 2,
    "cold_workers": 3,
    "max_workers": 10,
    "test_workers": 1,
    "gpu_ram": 16,
    "autoscaler_instance": "prod",
    "docker_login_user": "myuser",
    "docker_login_pass": "<string>",
    "docker_login_repo": "docker.io",
    "template_hash": "abc123def456",
    "template_id": 999,
    "launch_args": "--image pytorch/pytorch:latest --disk 20",
    "endpoint_name": "my-inference-endpoint",
    "endpoint_id": 42,
    "search_params": "gpu_ram>=16 num_gpus=1"
}
headers = {
    "Authorization": "Bearer <token>",
    "Content-Type": "application/json"
}

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

print(response.text)

const options = {
  method: 'POST',
  headers: {Authorization: 'Bearer <token>', 'Content-Type': 'application/json'},
  body: JSON.stringify({
    min_load: 0.1,
    min_cold_load: 0,
    target_util: 0.8,
    cold_mult: 2,
    cold_workers: 3,
    max_workers: 10,
    test_workers: 1,
    gpu_ram: 16,
    autoscaler_instance: 'prod',
    docker_login_user: 'myuser',
    docker_login_pass: '<string>',
    docker_login_repo: 'docker.io',
    template_hash: 'abc123def456',
    template_id: 999,
    launch_args: '--image pytorch/pytorch:latest --disk 20',
    endpoint_name: 'my-inference-endpoint',
    endpoint_id: 42,
    search_params: 'gpu_ram>=16 num_gpus=1'
  })
};

fetch('https://console.vast.ai/api/v0/autojobs/', 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://console.vast.ai/api/v0/autojobs/",
  CURLOPT_RETURNTRANSFER => true,
  CURLOPT_ENCODING => "",
  CURLOPT_MAXREDIRS => 10,
  CURLOPT_TIMEOUT => 30,
  CURLOPT_HTTP_VERSION => CURL_HTTP_VERSION_1_1,
  CURLOPT_CUSTOMREQUEST => "POST",
  CURLOPT_POSTFIELDS => json_encode([
    'min_load' => 0.1,
    'min_cold_load' => 0,
    'target_util' => 0.8,
    'cold_mult' => 2,
    'cold_workers' => 3,
    'max_workers' => 10,
    'test_workers' => 1,
    'gpu_ram' => 16,
    'autoscaler_instance' => 'prod',
    'docker_login_user' => 'myuser',
    'docker_login_pass' => '<string>',
    'docker_login_repo' => 'docker.io',
    'template_hash' => 'abc123def456',
    'template_id' => 999,
    'launch_args' => '--image pytorch/pytorch:latest --disk 20',
    'endpoint_name' => 'my-inference-endpoint',
    'endpoint_id' => 42,
    'search_params' => 'gpu_ram>=16 num_gpus=1'
  ]),
  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://console.vast.ai/api/v0/autojobs/"

	payload := strings.NewReader("{\n  \"min_load\": 0.1,\n  \"min_cold_load\": 0,\n  \"target_util\": 0.8,\n  \"cold_mult\": 2,\n  \"cold_workers\": 3,\n  \"max_workers\": 10,\n  \"test_workers\": 1,\n  \"gpu_ram\": 16,\n  \"autoscaler_instance\": \"prod\",\n  \"docker_login_user\": \"myuser\",\n  \"docker_login_pass\": \"<string>\",\n  \"docker_login_repo\": \"docker.io\",\n  \"template_hash\": \"abc123def456\",\n  \"template_id\": 999,\n  \"launch_args\": \"--image pytorch/pytorch:latest --disk 20\",\n  \"endpoint_name\": \"my-inference-endpoint\",\n  \"endpoint_id\": 42,\n  \"search_params\": \"gpu_ram>=16 num_gpus=1\"\n}")

	req, _ := http.NewRequest("POST", 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.post("https://console.vast.ai/api/v0/autojobs/")
  .header("Authorization", "Bearer <token>")
  .header("Content-Type", "application/json")
  .body("{\n  \"min_load\": 0.1,\n  \"min_cold_load\": 0,\n  \"target_util\": 0.8,\n  \"cold_mult\": 2,\n  \"cold_workers\": 3,\n  \"max_workers\": 10,\n  \"test_workers\": 1,\n  \"gpu_ram\": 16,\n  \"autoscaler_instance\": \"prod\",\n  \"docker_login_user\": \"myuser\",\n  \"docker_login_pass\": \"<string>\",\n  \"docker_login_repo\": \"docker.io\",\n  \"template_hash\": \"abc123def456\",\n  \"template_id\": 999,\n  \"launch_args\": \"--image pytorch/pytorch:latest --disk 20\",\n  \"endpoint_name\": \"my-inference-endpoint\",\n  \"endpoint_id\": 42,\n  \"search_params\": \"gpu_ram>=16 num_gpus=1\"\n}")
  .asString();

require 'uri'
require 'net/http'

url = URI("https://console.vast.ai/api/v0/autojobs/")

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

request = Net::HTTP::Post.new(url)
request["Authorization"] = 'Bearer <token>'
request["Content-Type"] = 'application/json'
request.body = "{\n  \"min_load\": 0.1,\n  \"min_cold_load\": 0,\n  \"target_util\": 0.8,\n  \"cold_mult\": 2,\n  \"cold_workers\": 3,\n  \"max_workers\": 10,\n  \"test_workers\": 1,\n  \"gpu_ram\": 16,\n  \"autoscaler_instance\": \"prod\",\n  \"docker_login_user\": \"myuser\",\n  \"docker_login_pass\": \"<string>\",\n  \"docker_login_repo\": \"docker.io\",\n  \"template_hash\": \"abc123def456\",\n  \"template_id\": 999,\n  \"launch_args\": \"--image pytorch/pytorch:latest --disk 20\",\n  \"endpoint_name\": \"my-inference-endpoint\",\n  \"endpoint_id\": 42,\n  \"search_params\": \"gpu_ram>=16 num_gpus=1\"\n}"

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

200

{
  "success": true,
  "id": 42
}

{
  "error": "<string>",
  "msg": "<string>"
}

{
  "error": "<string>",
  "msg": "<string>"
}

Authorizations

Authorization · string · header · required

API key must be provided in the Authorization header

Body

application/json

min_load · number

Minimum load threshold below which the autoscaler keeps warm workers running.

Example:

0.1

min_cold_load · number

Minimum baseline of cold (standby) load capacity the endpoint should maintain; applied when the endpoint is auto-created via endpoint_name.

Example:

0

target_util · number

Target utilization ratio (0-1) that the autoscaler aims to maintain across workers.

Example:

0.8

cold_mult · number

Multiplier controlling how many cold-standby workers are maintained relative to active workers.

Example:

2

cold_workers · integer

Fixed number of cold (standby) workers to keep available; must be <= max_workers.

Example:

3

max_workers · integer

Upper bound on the total number of workers this workergroup may scale up to.

Example:

10

test_workers · integer

Number of workers reserved for testing/canary use.

Example:

1

gpu_ram · number

Minimum GPU RAM in GB to target when selecting machines for this workergroup.

Example:

16

autoscaler_instance · string

Autoscaler deployment environment to use (e.g. 'prod', 'dev').

Example:

"prod"

docker_login_user · string

Docker registry username for pulling private images.

Example:

"myuser"

docker_login_pass · string

Docker registry password for pulling private images.

docker_login_repo · string

Docker registry hostname/repo to authenticate against.

Example:

"docker.io"

template_hash · string

Hash ID of a saved template to use for worker launch configuration.

Example:

"abc123def456"

template_id · integer

Numeric ID of a saved template to use for worker launch configuration.

Example:

999

launch_args · string

Additional CLI-style arguments to pass when launching worker instances.

Example:

"--image pytorch/pytorch:latest --disk 20"

endpoint_name · string

Human-readable name identifying the target endpoint; if no endpoint with this name exists, one is automatically created.

Example:

"my-inference-endpoint"

endpoint_id · integer

Numeric ID of an existing endpoint to attach this workergroup to; required when endpoint_name is not provided.

Example:

42

search_params · string

GPU offer filter expression (e.g. 'gpu_ram>=16 num_gpus=1') that the autoscaler uses to find machines for workers.

Example:

"gpu_ram>=16 num_gpus=1"

Response

200

application/json

Workergroup created; returns {success: true, id: <workergroup_id>}

success · boolean

Always true on success.

id · integer

ID of the newly created workergroup.