POST
/
gpu-instance
/
openapi
/
v1
/
gpu
/
instance
/
create

​
Request Headers

​
Content-Type
string
required

Enum: application/json

​
Authorization
string
required

Bearer authentication format, for example: Bearer {{API Key}}.

​
Request Body

​
name
string
required

Instance name.

​
productId
string
required

ID of the product used to deploy the instance.

​
gpuNum
integer
required

Number of GPUs allocated to the instance.

​
cpuNum
integer

Number of vCPU cores allocated to the instance.

​
memory
integer

Memory allocated to the instance (GB).

​
diskSize
integer
required

Disk space allocated to each instance (GB).

​
rootfsSize
integer
required

Root filesystem storage (GB).

​
imageUrl
string
required

Docker image URL to initialize the instance.

​
imageAuth
string

Docker image registry credentials in username:password format.

​
ports
string
required

Exposed ports.

​
envs
object[]

Environment variables.

​
tools
object[]

Official docker image built-in tools.

​
command
string
required

Startup command for the instance.

​
clusterId
string

ID of the cluster where the instance will be deployed.

​
localStorageMountPoint
string

Mount point for the local storage.

​
networkStorages
object[]

Cloud Mount Configuration (supports up to 30 cloud storage mounts)

​
networkStorageId
string

ID of the network storage.

​
networkStorageMountPoint
string

Mount point for the network storage.

​
networkId
string

VPC Network ID; leave empty if not using a VPC network.

​
kind
string
required

Instance TypeEnum: gpu

​
Response

​
id
string

ID of the created instance.

​
Example

Request

curl --request POST \
  --url 'https://api.novita.ai/gpu-instance/openapi/v1/gpu/instance/create' \\
  --header 'Authorization: Bearer {{API Key}}' \\
  --header 'Content-Type: application/json' \\
  --data '{
  "name": "gpu-test",                
  "productId": "26",                      
  "gpuNum": 1,                           
  "cpuNum": 0,                           
  "memory": 0,                        
  "diskSize": 60,                       
  "rootfsSize": 30,                       
  "imageUrl": "infrai/pytorch:2.2.1",     
  "imageAuth": "",                        
  "ports": "8080/tcp,8081/http",                  
  "envs": [                             
      {
          "key": "test",
          "value": "test"
      }
  ],
  "tools": [
      {
          "name": "Jupyter",
          "port": "8888",
          "type": "http"
      }
  ],
  "command": "",                         
  "clusterId": "",                            
  "localStorageMountPoint": "/workspace",
  "networkStorages": [
      {
          "Id": "e797ce8f-f4ff-4b3b-b7a8-6972d96f46f2",
          "mountPoint": "/network_0"
      },
      {
          "Id": "b924cdf5-82ec-4903-bdb2-74a03f2b0ae7",
          "mountPoint": "/network_1"
      }
  ],
  "networkId": "18f1ff05-1370-45ce-a4c6-b58d5e8d547e", 
  "kind": "gpu"
}'

Response

{
  "id": "<string>"
}

Was this page helpful?

List ClustersDelete Instance