OpenAI



Azure OpenAI Service provides REST API access to OpenAI's powerful language models including GPT-4, GPT-3, Codex, and Embeddings models.


Example yaml file:


apiVersion: azure.boundless.software/v1alpha1
kind: OpenAI
metadata:
  name: openai-sample
spec:
  provider:
    region: "eastus"
  resourceGroup: "my-resource-group"
  deployments:
    gpt4:
      version: "0301"
      model: "gpt-4"
      type: "model"
      tokenRateLimit: 5000
    embedding:
      version: "2"
      model: "text-embedding-ada-002"
      type: "embedding"
  networking:
    privateAccessOnly: true
    ipRules:
      - "203.0.113.0/24"
    subnetIDs:
      - "/subscriptions/00000000-0000-0000-0000-000000000000/resourceGroups/myResourceGroup/providers/Microsoft.Network/virtualNetworks/myVNet/subnets/mySubnet"
  privateEndpoints:
    - subnetID: "/subscriptions/00000000-0000-0000-0000-000000000000/resourceGroups/myResourceGroup/providers/Microsoft.Network/virtualNetworks/myVNet/subnets/mySubnet"
      location: "eastus"
  tags:
    environment: "production"
    department: "ai"

Inputs


NameTypeDescription
provider*AzureProviderSpecThe provider for the OpenAI resource.
resourceGroup*stringThe resource group for the OpenAI resource.
deploymentsmap[string]DeploymentSpecThe deployments for the OpenAI resource.
networkingNetworkingSpecThe networking configuration for the OpenAI resource.
privateEndpoints[]PrivateEndpointSpecThe private endpoints for the OpenAI resource.
tagsmap[string]*stringTags to apply to the OpenAI resource.

NetworkingSpec


NameTypeDescription
privateAccessOnlyboolWhether to restrict access to private networks only.
ipRules[]stringThe IP ranges allowed to access the OpenAI resource.
subnetIDs[]stringThe subnet IDs for the OpenAI resource to be accessed from.

DeploymentSpec


NameTypeDescription
versionstringThe version of the deployment.
modelstringThe model of the deployment.
typestringThe type of the deployment.
tokenRateLimitint32The token rate limit of the deployment.

PrivateEndpointSpec


NameTypeDescription
subnetID*stringThe subnet ID for the private endpoint.
location*stringThe location of the subnet.

DeploymentStatus


NameTypeDescription
conditionmetav1.ConditionThe condition of the deployment.
idstringThe ID of the deployment.
versionstringThe version of the deployment.
modelstringThe model of the deployment.
typestringThe type of the deployment.
tokenRateLimitint32The token rate limit of the deployment.

Outputs


NameTypeDescription
conditionmetav1.ConditionCurrent condition of the OpenAI resource.
resourceGroupstringThe resource group of the OpenAI resource.
deploymentsmap[string]DeploymentStatusStatus of the deployments.
privateEndpointsmap[string]PrivateEndpointStatusStatus of the private endpoints.
networkingNetworkingStatusStatus of the networking configuration.
locationstringThe location of the OpenAI resource.
namestringThe name of the OpenAI resource.
idstringThe ID of the OpenAI resource.
endpointstringThe endpoints of the OpenAI resource.
keysKeysStatusThe keys of the OpenAI resource.

NetworkingStatus


NameTypeDescription
privateAccessOnlyboolWhether access is restricted to private networks.
ipRules[]stringThe IP ranges allowed to access the resource.
subnetIDs[]stringThe subnet IDs for resource access.

KeysStatus


NameTypeDescription
primaryKeystringThe primary key of the OpenAI resource.
secondaryKeystringThe secondary key of the OpenAI resource.