Create Instance API

Provision a new GPU instance on a specific provider with custom resources and a docker image.

Request Body

provider_namestringrequired

Name of the provider (e.g., 'novita', 'verda', 'runpod').

instance_namestringrequired

User-defined name for the instance.

gpu_idstringrequired

Identifier for the requested GPU type.

gpu_numnumberrequired

Quantity of GPUs to provision.

disk_sizenumberrequired

Size of the root disk in GB.

image_urlstringrequired

Docker image URL to deploy.

regionstringrequired

Region identifier (e.g., 'us-east', 'eu-west').

Example Request Body

{
  "provider_name": "novita",
  "instance_name": "my-gpu-instance",
  "gpu_id": "nvidia_a100_80gb",
  "gpu_num": 1,
  "disk_size": 100,
  "image_url": "pytorch/pytorch:latest",
  "region": "us-west-1"
}

Error Status Codes

Code	Status	Description
200	OK	Instance provisioned successfully.
400	Bad Request	Invalid request body or unsupported provider.
401	Unauthorized	Missing or invalid Authorization header / Bearer token.
403	Forbidden	Insufficient credit balance or provider validation failed.
422	Unprocessable Entity	Request body failed schema validation.
500	Internal Server Error	Unexpected error during instance provisioning.

Create Instance

Request Body

Example Request Body

Error Status Codes

Response