{ "swagger": "2.0", "info": { "title": "ContainerServiceAIManagerClient", "version": "2026-05-02-preview", "description": "Azure Kubernetes AI Manager api client.", "x-typespec-generated": [ { "emitter": "@azure-tools/typespec-autorest" } ] }, "schemes": [ "https" ], "host": "management.azure.com", "produces": [ "application/json" ], "consumes": [ "application/json" ], "security": [ { "azure_auth": [ "user_impersonation" ] } ], "securityDefinitions": { "azure_auth": { "type": "oauth2", "description": "Azure Active Directory OAuth2 Flow.", "flow": "implicit", "authorizationUrl": "https://login.microsoftonline.com/common/oauth2/authorize", "scopes": { "user_impersonation": "impersonate your user account" } } }, "tags": [ { "name": "Operations" }, { "name": "AIManagers" }, { "name": "AIManagerNamespaces" }, { "name": "AIModels" }, { "name": "ModelSources" }, { "name": "ModelDeployments" } ], "paths": { "/providers/Microsoft.ContainerService/operations": { "get": { "operationId": "Operations_List", "tags": [ "Operations" ], "description": "List the operations for the provider", "parameters": [ { "$ref": "../../../../../../common-types/resource-management/v6/types.json#/parameters/ApiVersionParameter" } ], "responses": { "200": { "description": "Azure operation completed successfully.", "schema": { "$ref": "../../../../../../common-types/resource-management/v6/types.json#/definitions/OperationListResult" } }, "default": { "description": "An unexpected error response.", "schema": { "$ref": "../../../../../../common-types/resource-management/v6/types.json#/definitions/ErrorResponse" } } }, "x-ms-examples": { "List the operations for the provider.": { "$ref": "./examples/Operations_List.json" } }, "x-ms-pageable": { "nextLinkName": "nextLink" } } }, "/subscriptions/{subscriptionId}/providers/Microsoft.ContainerService/aiManagers": { "get": { "operationId": "AIManagers_ListBySubscription", "tags": [ "AIManagers" ], "description": "List AIManager resources by subscription ID", "parameters": [ { "$ref": "../../../../../../common-types/resource-management/v6/types.json#/parameters/ApiVersionParameter" }, { "$ref": "../../../../../../common-types/resource-management/v6/types.json#/parameters/SubscriptionIdParameter" } ], "responses": { "200": { "description": "Azure operation completed successfully.", "schema": { "$ref": "#/definitions/AIManagerListResult" } }, "default": { "description": "An unexpected error response.", "schema": { "$ref": "../../../../../../common-types/resource-management/v6/types.json#/definitions/ErrorResponse" } } }, "x-ms-examples": { "Lists AI Manager resources by subscription.": { "$ref": "./examples/AIManagers_ListBySubscription.json" } }, "x-ms-pageable": { "nextLinkName": "nextLink" } } }, "/subscriptions/{subscriptionId}/providers/Microsoft.ContainerService/aiModels": { "get": { "operationId": "AIModels_ListBySubscription", "tags": [ "AIModels" ], "description": "List AIModel resources by subscription ID", "parameters": [ { "$ref": "../../../../../../common-types/resource-management/v6/types.json#/parameters/ApiVersionParameter" }, { "$ref": "../../../../../../common-types/resource-management/v6/types.json#/parameters/SubscriptionIdParameter" } ], "responses": { "200": { "description": "Azure operation completed successfully.", "schema": { "$ref": "#/definitions/AIModelListResult" } }, "default": { "description": "An unexpected error response.", "schema": { "$ref": "../../../../../../common-types/resource-management/v6/types.json#/definitions/ErrorResponse" } } }, "x-ms-examples": { "AIModels_ListBySubscription_MaximumSet": { "$ref": "./examples/AIModels_ListBySubscription.json" } }, "x-ms-pageable": { "nextLinkName": "nextLink" } } }, "/subscriptions/{subscriptionId}/resourceGroups/{resourceGroupName}/providers/Microsoft.ContainerService/aiManagers": { "get": { "operationId": "AIManagers_ListByResourceGroup", "tags": [ "AIManagers" ], "description": "List AIManager resources by resource group", "parameters": [ { "$ref": "../../../../../../common-types/resource-management/v6/types.json#/parameters/ApiVersionParameter" }, { "$ref": "../../../../../../common-types/resource-management/v6/types.json#/parameters/SubscriptionIdParameter" }, { "$ref": "../../../../../../common-types/resource-management/v6/types.json#/parameters/ResourceGroupNameParameter" } ], "responses": { "200": { "description": "Azure operation completed successfully.", "schema": { "$ref": "#/definitions/AIManagerListResult" } }, "default": { "description": "An unexpected error response.", "schema": { "$ref": "../../../../../../common-types/resource-management/v6/types.json#/definitions/ErrorResponse" } } }, "x-ms-examples": { "Lists AI Manager resources by resource group.": { "$ref": "./examples/AIManagers_ListByResourceGroup.json" } }, "x-ms-pageable": { "nextLinkName": "nextLink" } } }, "/subscriptions/{subscriptionId}/resourceGroups/{resourceGroupName}/providers/Microsoft.ContainerService/aiManagers/{aiManagerName}": { "get": { "operationId": "AIManagers_Get", "tags": [ "AIManagers" ], "description": "Get a AIManager", "parameters": [ { "$ref": "../../../../../../common-types/resource-management/v6/types.json#/parameters/ApiVersionParameter" }, { "$ref": "../../../../../../common-types/resource-management/v6/types.json#/parameters/SubscriptionIdParameter" }, { "$ref": "../../../../../../common-types/resource-management/v6/types.json#/parameters/ResourceGroupNameParameter" }, { "name": "aiManagerName", "in": "path", "description": "The name of the AI Manager resource.", "required": true, "type": "string", "minLength": 1, "maxLength": 63, "pattern": "^[a-z0-9]([-a-z0-9]*[a-z0-9])?$" } ], "responses": { "200": { "description": "Azure operation completed successfully.", "schema": { "$ref": "#/definitions/AIManager" } }, "default": { "description": "An unexpected error response.", "schema": { "$ref": "../../../../../../common-types/resource-management/v6/types.json#/definitions/ErrorResponse" } } }, "x-ms-examples": { "Gets an AI Manager resource.": { "$ref": "./examples/AIManagers_Get.json" } } }, "put": { "operationId": "AIManagers_CreateOrUpdate", "tags": [ "AIManagers" ], "description": "Create a AIManager", "parameters": [ { "$ref": "../../../../../../common-types/resource-management/v6/types.json#/parameters/ApiVersionParameter" }, { "$ref": "../../../../../../common-types/resource-management/v6/types.json#/parameters/SubscriptionIdParameter" }, { "$ref": "../../../../../../common-types/resource-management/v6/types.json#/parameters/ResourceGroupNameParameter" }, { "name": "If-Match", "in": "header", "description": "The request should only proceed if an entity matches this string.", "required": false, "type": "string", "x-ms-client-name": "ifMatch" }, { "name": "If-None-Match", "in": "header", "description": "The request should only proceed if no entity matches this string.", "required": false, "type": "string", "x-ms-client-name": "ifNoneMatch" }, { "name": "aiManagerName", "in": "path", "description": "The name of the AI Manager resource.", "required": true, "type": "string", "minLength": 1, "maxLength": 63, "pattern": "^[a-z0-9]([-a-z0-9]*[a-z0-9])?$" }, { "name": "resource", "in": "body", "description": "Resource create parameters.", "required": true, "schema": { "$ref": "#/definitions/AIManager" } } ], "responses": { "200": { "description": "Resource 'AIManager' update operation succeeded", "schema": { "$ref": "#/definitions/AIManager" } }, "201": { "description": "Resource 'AIManager' create operation succeeded", "schema": { "$ref": "#/definitions/AIManager" }, "headers": { "Azure-AsyncOperation": { "type": "string", "description": "A link to the status monitor" }, "Retry-After": { "type": "integer", "format": "int32", "description": "The Retry-After header can indicate how long the client should wait before polling the operation status." } } }, "default": { "description": "An unexpected error response.", "schema": { "$ref": "../../../../../../common-types/resource-management/v6/types.json#/definitions/ErrorResponse" } } }, "x-ms-examples": { "Creates or updates an AI Manager resource.": { "$ref": "./examples/AIManagers_CreateOrUpdate.json" } }, "x-ms-long-running-operation-options": { "final-state-via": "azure-async-operation" }, "x-ms-long-running-operation": true }, "patch": { "operationId": "AIManagers_Update", "tags": [ "AIManagers" ], "description": "Update a AIManager", "parameters": [ { "$ref": "../../../../../../common-types/resource-management/v6/types.json#/parameters/ApiVersionParameter" }, { "$ref": "../../../../../../common-types/resource-management/v6/types.json#/parameters/SubscriptionIdParameter" }, { "$ref": "../../../../../../common-types/resource-management/v6/types.json#/parameters/ResourceGroupNameParameter" }, { "name": "If-Match", "in": "header", "description": "The request should only proceed if an entity matches this string.", "required": false, "type": "string", "x-ms-client-name": "ifMatch" }, { "name": "aiManagerName", "in": "path", "description": "The name of the AI Manager resource.", "required": true, "type": "string", "minLength": 1, "maxLength": 63, "pattern": "^[a-z0-9]([-a-z0-9]*[a-z0-9])?$" }, { "name": "properties", "in": "body", "description": "The resource properties to be updated.", "required": true, "schema": { "$ref": "#/definitions/AIManagerPatch" } } ], "responses": { "200": { "description": "Azure operation completed successfully.", "schema": { "$ref": "#/definitions/AIManager" } }, "default": { "description": "An unexpected error response.", "schema": { "$ref": "../../../../../../common-types/resource-management/v6/types.json#/definitions/ErrorResponse" } } }, "x-ms-examples": { "Updates an AI Manager resource.": { "$ref": "./examples/AIManagers_Update.json" } } }, "delete": { "operationId": "AIManagers_Delete", "tags": [ "AIManagers" ], "description": "Delete a AIManager", "parameters": [ { "$ref": "../../../../../../common-types/resource-management/v6/types.json#/parameters/ApiVersionParameter" }, { "$ref": "../../../../../../common-types/resource-management/v6/types.json#/parameters/SubscriptionIdParameter" }, { "$ref": "../../../../../../common-types/resource-management/v6/types.json#/parameters/ResourceGroupNameParameter" }, { "name": "If-Match", "in": "header", "description": "The request should only proceed if an entity matches this string.", "required": false, "type": "string", "x-ms-client-name": "ifMatch" }, { "name": "aiManagerName", "in": "path", "description": "The name of the AI Manager resource.", "required": true, "type": "string", "minLength": 1, "maxLength": 63, "pattern": "^[a-z0-9]([-a-z0-9]*[a-z0-9])?$" } ], "responses": { "202": { "description": "Resource deletion accepted.", "headers": { "Location": { "type": "string", "description": "The Location header contains the URL where the status of the long running operation can be checked." }, "Retry-After": { "type": "integer", "format": "int32", "description": "The Retry-After header can indicate how long the client should wait before polling the operation status." } } }, "204": { "description": "Resource does not exist." }, "default": { "description": "An unexpected error response.", "schema": { "$ref": "../../../../../../common-types/resource-management/v6/types.json#/definitions/ErrorResponse" } } }, "x-ms-examples": { "Deletes an AI Manager resource.": { "$ref": "./examples/AIManagers_Delete.json" } }, "x-ms-long-running-operation-options": { "final-state-via": "location" }, "x-ms-long-running-operation": true } }, "/subscriptions/{subscriptionId}/resourceGroups/{resourceGroupName}/providers/Microsoft.ContainerService/aiManagers/{aiManagerName}/listCredential": { "post": { "operationId": "AIManagers_ListCredential", "tags": [ "AIManagers" ], "description": "Lists the credentials of an AI Manager.", "parameters": [ { "$ref": "../../../../../../common-types/resource-management/v6/types.json#/parameters/ApiVersionParameter" }, { "$ref": "../../../../../../common-types/resource-management/v6/types.json#/parameters/SubscriptionIdParameter" }, { "$ref": "../../../../../../common-types/resource-management/v6/types.json#/parameters/ResourceGroupNameParameter" }, { "name": "aiManagerName", "in": "path", "description": "The name of the AI Manager resource.", "required": true, "type": "string", "minLength": 1, "maxLength": 63, "pattern": "^[a-z0-9]([-a-z0-9]*[a-z0-9])?$" } ], "responses": { "200": { "description": "Azure operation completed successfully.", "schema": { "$ref": "#/definitions/CredentialResults" } }, "default": { "description": "An unexpected error response.", "schema": { "$ref": "../../../../../../common-types/resource-management/v6/types.json#/definitions/ErrorResponse" } } }, "x-ms-examples": { "Lists the credentials of an AI Manager.": { "$ref": "./examples/AIManagers_ListCredential.json" } } } }, "/subscriptions/{subscriptionId}/resourceGroups/{resourceGroupName}/providers/Microsoft.ContainerService/aiManagers/{aiManagerName}/modelSources": { "get": { "operationId": "ModelSources_ListByAIManager", "tags": [ "ModelSources" ], "description": "List ModelSource resources by AIManager", "parameters": [ { "$ref": "../../../../../../common-types/resource-management/v6/types.json#/parameters/ApiVersionParameter" }, { "$ref": "../../../../../../common-types/resource-management/v6/types.json#/parameters/SubscriptionIdParameter" }, { "$ref": "../../../../../../common-types/resource-management/v6/types.json#/parameters/ResourceGroupNameParameter" }, { "name": "aiManagerName", "in": "path", "description": "The name of the AI Manager resource.", "required": true, "type": "string", "minLength": 1, "maxLength": 63, "pattern": "^[a-z0-9]([-a-z0-9]*[a-z0-9])?$" } ], "responses": { "200": { "description": "Azure operation completed successfully.", "schema": { "$ref": "#/definitions/ModelSourceListResult" } }, "default": { "description": "An unexpected error response.", "schema": { "$ref": "../../../../../../common-types/resource-management/v6/types.json#/definitions/ErrorResponse" } } }, "x-ms-examples": { "ModelSources_ListByAIManager_MaximumSet": { "$ref": "./examples/ModelSources_ListByAIManager.json" } }, "x-ms-pageable": { "nextLinkName": "nextLink" } } }, "/subscriptions/{subscriptionId}/resourceGroups/{resourceGroupName}/providers/Microsoft.ContainerService/aiManagers/{aiManagerName}/modelSources/{modelSourceName}": { "get": { "operationId": "ModelSources_Get", "tags": [ "ModelSources" ], "description": "Get a ModelSource", "parameters": [ { "$ref": "../../../../../../common-types/resource-management/v6/types.json#/parameters/ApiVersionParameter" }, { "$ref": "../../../../../../common-types/resource-management/v6/types.json#/parameters/SubscriptionIdParameter" }, { "$ref": "../../../../../../common-types/resource-management/v6/types.json#/parameters/ResourceGroupNameParameter" }, { "name": "aiManagerName", "in": "path", "description": "The name of the AI Manager resource.", "required": true, "type": "string", "minLength": 1, "maxLength": 63, "pattern": "^[a-z0-9]([-a-z0-9]*[a-z0-9])?$" }, { "name": "modelSourceName", "in": "path", "description": "The name of the model source resource.", "required": true, "type": "string", "minLength": 1, "maxLength": 63, "pattern": "^[a-z0-9]([-a-z0-9]*[a-z0-9])?$" } ], "responses": { "200": { "description": "Azure operation completed successfully.", "schema": { "$ref": "#/definitions/ModelSource" } }, "default": { "description": "An unexpected error response.", "schema": { "$ref": "../../../../../../common-types/resource-management/v6/types.json#/definitions/ErrorResponse" } } }, "x-ms-examples": { "ModelSources_Get_MaximumSet": { "$ref": "./examples/ModelSources_Get.json" } } }, "put": { "operationId": "ModelSources_CreateOrUpdate", "tags": [ "ModelSources" ], "description": "Create a ModelSource", "parameters": [ { "$ref": "../../../../../../common-types/resource-management/v6/types.json#/parameters/ApiVersionParameter" }, { "$ref": "../../../../../../common-types/resource-management/v6/types.json#/parameters/SubscriptionIdParameter" }, { "$ref": "../../../../../../common-types/resource-management/v6/types.json#/parameters/ResourceGroupNameParameter" }, { "name": "If-Match", "in": "header", "description": "The request should only proceed if an entity matches this string.", "required": false, "type": "string", "x-ms-client-name": "ifMatch" }, { "name": "If-None-Match", "in": "header", "description": "The request should only proceed if no entity matches this string.", "required": false, "type": "string", "x-ms-client-name": "ifNoneMatch" }, { "name": "aiManagerName", "in": "path", "description": "The name of the AI Manager resource.", "required": true, "type": "string", "minLength": 1, "maxLength": 63, "pattern": "^[a-z0-9]([-a-z0-9]*[a-z0-9])?$" }, { "name": "modelSourceName", "in": "path", "description": "The name of the model source resource.", "required": true, "type": "string", "minLength": 1, "maxLength": 63, "pattern": "^[a-z0-9]([-a-z0-9]*[a-z0-9])?$" }, { "name": "resource", "in": "body", "description": "Resource create parameters.", "required": true, "schema": { "$ref": "#/definitions/ModelSource" } } ], "responses": { "200": { "description": "Resource 'ModelSource' update operation succeeded", "schema": { "$ref": "#/definitions/ModelSource" } }, "201": { "description": "Resource 'ModelSource' create operation succeeded", "schema": { "$ref": "#/definitions/ModelSource" } }, "default": { "description": "An unexpected error response.", "schema": { "$ref": "../../../../../../common-types/resource-management/v6/types.json#/definitions/ErrorResponse" } } }, "x-ms-examples": { "ModelSources_CreateOrUpdate_MaximumSet": { "$ref": "./examples/ModelSources_CreateOrUpdate.json" } } }, "patch": { "operationId": "ModelSources_Update", "tags": [ "ModelSources" ], "description": "Update a ModelSource", "parameters": [ { "$ref": "../../../../../../common-types/resource-management/v6/types.json#/parameters/ApiVersionParameter" }, { "$ref": "../../../../../../common-types/resource-management/v6/types.json#/parameters/SubscriptionIdParameter" }, { "$ref": "../../../../../../common-types/resource-management/v6/types.json#/parameters/ResourceGroupNameParameter" }, { "name": "If-Match", "in": "header", "description": "The request should only proceed if an entity matches this string.", "required": false, "type": "string", "x-ms-client-name": "ifMatch" }, { "name": "aiManagerName", "in": "path", "description": "The name of the AI Manager resource.", "required": true, "type": "string", "minLength": 1, "maxLength": 63, "pattern": "^[a-z0-9]([-a-z0-9]*[a-z0-9])?$" }, { "name": "modelSourceName", "in": "path", "description": "The name of the model source resource.", "required": true, "type": "string", "minLength": 1, "maxLength": 63, "pattern": "^[a-z0-9]([-a-z0-9]*[a-z0-9])?$" }, { "name": "properties", "in": "body", "description": "The resource properties to be updated.", "required": true, "schema": { "$ref": "#/definitions/ModelSourcePatch" } } ], "responses": { "200": { "description": "Azure operation completed successfully.", "schema": { "$ref": "#/definitions/ModelSource" } }, "default": { "description": "An unexpected error response.", "schema": { "$ref": "../../../../../../common-types/resource-management/v6/types.json#/definitions/ErrorResponse" } } }, "x-ms-examples": { "ModelSources_Update_MaximumSet": { "$ref": "./examples/ModelSources_Update.json" } } }, "delete": { "operationId": "ModelSources_Delete", "tags": [ "ModelSources" ], "description": "Delete a ModelSource", "parameters": [ { "$ref": "../../../../../../common-types/resource-management/v6/types.json#/parameters/ApiVersionParameter" }, { "$ref": "../../../../../../common-types/resource-management/v6/types.json#/parameters/SubscriptionIdParameter" }, { "$ref": "../../../../../../common-types/resource-management/v6/types.json#/parameters/ResourceGroupNameParameter" }, { "name": "If-Match", "in": "header", "description": "The request should only proceed if an entity matches this string.", "required": false, "type": "string", "x-ms-client-name": "ifMatch" }, { "name": "aiManagerName", "in": "path", "description": "The name of the AI Manager resource.", "required": true, "type": "string", "minLength": 1, "maxLength": 63, "pattern": "^[a-z0-9]([-a-z0-9]*[a-z0-9])?$" }, { "name": "modelSourceName", "in": "path", "description": "The name of the model source resource.", "required": true, "type": "string", "minLength": 1, "maxLength": 63, "pattern": "^[a-z0-9]([-a-z0-9]*[a-z0-9])?$" } ], "responses": { "200": { "description": "Resource deleted successfully." }, "204": { "description": "Resource does not exist." }, "default": { "description": "An unexpected error response.", "schema": { "$ref": "../../../../../../common-types/resource-management/v6/types.json#/definitions/ErrorResponse" } } }, "x-ms-examples": { "ModelSources_Delete_MaximumSet": { "$ref": "./examples/ModelSources_Delete.json" } } } }, "/subscriptions/{subscriptionId}/resourceGroups/{resourceGroupName}/providers/Microsoft.ContainerService/aiManagers/{aiManagerName}/namespaces": { "get": { "operationId": "AIManagerNamespaces_ListByAIManager", "tags": [ "AIManagerNamespaces" ], "description": "List AIManagerNamespace resources by AIManager", "parameters": [ { "$ref": "../../../../../../common-types/resource-management/v6/types.json#/parameters/ApiVersionParameter" }, { "$ref": "../../../../../../common-types/resource-management/v6/types.json#/parameters/SubscriptionIdParameter" }, { "$ref": "../../../../../../common-types/resource-management/v6/types.json#/parameters/ResourceGroupNameParameter" }, { "name": "aiManagerName", "in": "path", "description": "The name of the AI Manager resource.", "required": true, "type": "string", "minLength": 1, "maxLength": 63, "pattern": "^[a-z0-9]([-a-z0-9]*[a-z0-9])?$" } ], "responses": { "200": { "description": "Azure operation completed successfully.", "schema": { "$ref": "#/definitions/AIManagerNamespaceListResult" } }, "default": { "description": "An unexpected error response.", "schema": { "$ref": "../../../../../../common-types/resource-management/v6/types.json#/definitions/ErrorResponse" } } }, "x-ms-examples": { "Lists AI Manager namespace resources by AI Manager.": { "$ref": "./examples/AIManagerNamespaces_ListByAIManager.json" } }, "x-ms-pageable": { "nextLinkName": "nextLink" } } }, "/subscriptions/{subscriptionId}/resourceGroups/{resourceGroupName}/providers/Microsoft.ContainerService/aiManagers/{aiManagerName}/namespaces/{namespaceName}": { "get": { "operationId": "AIManagerNamespaces_Get", "tags": [ "AIManagerNamespaces" ], "description": "Get a AIManagerNamespace", "parameters": [ { "$ref": "../../../../../../common-types/resource-management/v6/types.json#/parameters/ApiVersionParameter" }, { "$ref": "../../../../../../common-types/resource-management/v6/types.json#/parameters/SubscriptionIdParameter" }, { "$ref": "../../../../../../common-types/resource-management/v6/types.json#/parameters/ResourceGroupNameParameter" }, { "name": "aiManagerName", "in": "path", "description": "The name of the AI Manager resource.", "required": true, "type": "string", "minLength": 1, "maxLength": 63, "pattern": "^[a-z0-9]([-a-z0-9]*[a-z0-9])?$" }, { "name": "namespaceName", "in": "path", "description": "The name of the AI Manager namespace resource.", "required": true, "type": "string", "minLength": 1, "maxLength": 63, "pattern": "^[a-z0-9]([-a-z0-9]*[a-z0-9])?$" } ], "responses": { "200": { "description": "Azure operation completed successfully.", "schema": { "$ref": "#/definitions/AIManagerNamespace" } }, "default": { "description": "An unexpected error response.", "schema": { "$ref": "../../../../../../common-types/resource-management/v6/types.json#/definitions/ErrorResponse" } } }, "x-ms-examples": { "Gets an AI Manager namespace resource.": { "$ref": "./examples/AIManagerNamespaces_Get.json" } } }, "put": { "operationId": "AIManagerNamespaces_CreateOrUpdate", "tags": [ "AIManagerNamespaces" ], "description": "Create a AIManagerNamespace", "parameters": [ { "$ref": "../../../../../../common-types/resource-management/v6/types.json#/parameters/ApiVersionParameter" }, { "$ref": "../../../../../../common-types/resource-management/v6/types.json#/parameters/SubscriptionIdParameter" }, { "$ref": "../../../../../../common-types/resource-management/v6/types.json#/parameters/ResourceGroupNameParameter" }, { "name": "If-Match", "in": "header", "description": "The request should only proceed if an entity matches this string.", "required": false, "type": "string", "x-ms-client-name": "ifMatch" }, { "name": "If-None-Match", "in": "header", "description": "The request should only proceed if no entity matches this string.", "required": false, "type": "string", "x-ms-client-name": "ifNoneMatch" }, { "name": "aiManagerName", "in": "path", "description": "The name of the AI Manager resource.", "required": true, "type": "string", "minLength": 1, "maxLength": 63, "pattern": "^[a-z0-9]([-a-z0-9]*[a-z0-9])?$" }, { "name": "namespaceName", "in": "path", "description": "The name of the AI Manager namespace resource.", "required": true, "type": "string", "minLength": 1, "maxLength": 63, "pattern": "^[a-z0-9]([-a-z0-9]*[a-z0-9])?$" }, { "name": "resource", "in": "body", "description": "Resource create parameters.", "required": true, "schema": { "$ref": "#/definitions/AIManagerNamespace" } } ], "responses": { "200": { "description": "Resource 'AIManagerNamespace' update operation succeeded", "schema": { "$ref": "#/definitions/AIManagerNamespace" } }, "201": { "description": "Resource 'AIManagerNamespace' create operation succeeded", "schema": { "$ref": "#/definitions/AIManagerNamespace" }, "headers": { "Azure-AsyncOperation": { "type": "string", "description": "A link to the status monitor" }, "Retry-After": { "type": "integer", "format": "int32", "description": "The Retry-After header can indicate how long the client should wait before polling the operation status." } } }, "default": { "description": "An unexpected error response.", "schema": { "$ref": "../../../../../../common-types/resource-management/v6/types.json#/definitions/ErrorResponse" } } }, "x-ms-examples": { "Creates or updates an AI Manager namespace resource.": { "$ref": "./examples/AIManagerNamespaces_CreateOrUpdate.json" } }, "x-ms-long-running-operation-options": { "final-state-via": "azure-async-operation" }, "x-ms-long-running-operation": true }, "delete": { "operationId": "AIManagerNamespaces_Delete", "tags": [ "AIManagerNamespaces" ], "description": "Delete a AIManagerNamespace", "parameters": [ { "$ref": "../../../../../../common-types/resource-management/v6/types.json#/parameters/ApiVersionParameter" }, { "$ref": "../../../../../../common-types/resource-management/v6/types.json#/parameters/SubscriptionIdParameter" }, { "$ref": "../../../../../../common-types/resource-management/v6/types.json#/parameters/ResourceGroupNameParameter" }, { "name": "If-Match", "in": "header", "description": "The request should only proceed if an entity matches this string.", "required": false, "type": "string", "x-ms-client-name": "ifMatch" }, { "name": "aiManagerName", "in": "path", "description": "The name of the AI Manager resource.", "required": true, "type": "string", "minLength": 1, "maxLength": 63, "pattern": "^[a-z0-9]([-a-z0-9]*[a-z0-9])?$" }, { "name": "namespaceName", "in": "path", "description": "The name of the AI Manager namespace resource.", "required": true, "type": "string", "minLength": 1, "maxLength": 63, "pattern": "^[a-z0-9]([-a-z0-9]*[a-z0-9])?$" } ], "responses": { "202": { "description": "Resource deletion accepted.", "headers": { "Location": { "type": "string", "description": "The Location header contains the URL where the status of the long running operation can be checked." }, "Retry-After": { "type": "integer", "format": "int32", "description": "The Retry-After header can indicate how long the client should wait before polling the operation status." } } }, "204": { "description": "Resource does not exist." }, "default": { "description": "An unexpected error response.", "schema": { "$ref": "../../../../../../common-types/resource-management/v6/types.json#/definitions/ErrorResponse" } } }, "x-ms-examples": { "Deletes an AI Manager namespace resource.": { "$ref": "./examples/AIManagerNamespaces_Delete.json" } }, "x-ms-long-running-operation-options": { "final-state-via": "location" }, "x-ms-long-running-operation": true } }, "/subscriptions/{subscriptionId}/resourceGroups/{resourceGroupName}/providers/Microsoft.ContainerService/aiManagers/{aiManagerName}/namespaces/{namespaceName}/listAccessInfo": { "post": { "operationId": "AIManagerNamespaces_ListAccessInfo", "tags": [ "AIManagerNamespaces" ], "description": "Returns the namespace-scoped LLM gateway endpoint and the API keys clients use to authenticate against it.", "parameters": [ { "$ref": "../../../../../../common-types/resource-management/v6/types.json#/parameters/ApiVersionParameter" }, { "$ref": "../../../../../../common-types/resource-management/v6/types.json#/parameters/SubscriptionIdParameter" }, { "$ref": "../../../../../../common-types/resource-management/v6/types.json#/parameters/ResourceGroupNameParameter" }, { "name": "aiManagerName", "in": "path", "description": "The name of the AI Manager resource.", "required": true, "type": "string", "minLength": 1, "maxLength": 63, "pattern": "^[a-z0-9]([-a-z0-9]*[a-z0-9])?$" }, { "name": "namespaceName", "in": "path", "description": "The name of the AI Manager namespace resource.", "required": true, "type": "string", "minLength": 1, "maxLength": 63, "pattern": "^[a-z0-9]([-a-z0-9]*[a-z0-9])?$" } ], "responses": { "200": { "description": "Azure operation completed successfully.", "schema": { "$ref": "#/definitions/NamespaceAccessInfo" } }, "default": { "description": "An unexpected error response.", "schema": { "$ref": "../../../../../../common-types/resource-management/v6/types.json#/definitions/ErrorResponse" } } }, "x-ms-examples": { "AIManagerNamespaces_ListAccessInfo_MaximumSet": { "$ref": "./examples/AIManagerNamespaces_ListAccessInfo.json" } } } }, "/subscriptions/{subscriptionId}/resourceGroups/{resourceGroupName}/providers/Microsoft.ContainerService/aiManagers/{aiManagerName}/namespaces/{namespaceName}/listCredential": { "post": { "operationId": "AIManagerNamespaces_ListCredential", "tags": [ "AIManagerNamespaces" ], "description": "Lists the credentials of an AI Manager namespace.", "parameters": [ { "$ref": "../../../../../../common-types/resource-management/v6/types.json#/parameters/ApiVersionParameter" }, { "$ref": "../../../../../../common-types/resource-management/v6/types.json#/parameters/SubscriptionIdParameter" }, { "$ref": "../../../../../../common-types/resource-management/v6/types.json#/parameters/ResourceGroupNameParameter" }, { "name": "aiManagerName", "in": "path", "description": "The name of the AI Manager resource.", "required": true, "type": "string", "minLength": 1, "maxLength": 63, "pattern": "^[a-z0-9]([-a-z0-9]*[a-z0-9])?$" }, { "name": "namespaceName", "in": "path", "description": "The name of the AI Manager namespace resource.", "required": true, "type": "string", "minLength": 1, "maxLength": 63, "pattern": "^[a-z0-9]([-a-z0-9]*[a-z0-9])?$" } ], "responses": { "200": { "description": "Azure operation completed successfully.", "schema": { "$ref": "#/definitions/CredentialResults" } }, "default": { "description": "An unexpected error response.", "schema": { "$ref": "../../../../../../common-types/resource-management/v6/types.json#/definitions/ErrorResponse" } } }, "x-ms-examples": { "Lists the credentials of an AI Manager namespace.": { "$ref": "./examples/AIManagerNamespaces_ListCredential.json" } } } }, "/subscriptions/{subscriptionId}/resourceGroups/{resourceGroupName}/providers/Microsoft.ContainerService/aiManagers/{aiManagerName}/namespaces/{namespaceName}/modelDeployments": { "get": { "operationId": "ModelDeployments_ListByAIManagerNamespace", "tags": [ "ModelDeployments" ], "description": "List ModelDeployment resources by AIManagerNamespace", "parameters": [ { "$ref": "../../../../../../common-types/resource-management/v6/types.json#/parameters/ApiVersionParameter" }, { "$ref": "../../../../../../common-types/resource-management/v6/types.json#/parameters/SubscriptionIdParameter" }, { "$ref": "../../../../../../common-types/resource-management/v6/types.json#/parameters/ResourceGroupNameParameter" }, { "name": "aiManagerName", "in": "path", "description": "The name of the AI Manager resource.", "required": true, "type": "string", "minLength": 1, "maxLength": 63, "pattern": "^[a-z0-9]([-a-z0-9]*[a-z0-9])?$" }, { "name": "namespaceName", "in": "path", "description": "The name of the AI Manager namespace resource.", "required": true, "type": "string", "minLength": 1, "maxLength": 63, "pattern": "^[a-z0-9]([-a-z0-9]*[a-z0-9])?$" } ], "responses": { "200": { "description": "Azure operation completed successfully.", "schema": { "$ref": "#/definitions/ModelDeploymentListResult" } }, "default": { "description": "An unexpected error response.", "schema": { "$ref": "../../../../../../common-types/resource-management/v6/types.json#/definitions/ErrorResponse" } } }, "x-ms-examples": { "ModelDeployments_ListByAIManagerNamespace_MaximumSet": { "$ref": "./examples/ModelDeployments_ListByAIManagerNamespace.json" } }, "x-ms-pageable": { "nextLinkName": "nextLink" } } }, "/subscriptions/{subscriptionId}/resourceGroups/{resourceGroupName}/providers/Microsoft.ContainerService/aiManagers/{aiManagerName}/namespaces/{namespaceName}/modelDeployments/{modelDeploymentName}": { "get": { "operationId": "ModelDeployments_Get", "tags": [ "ModelDeployments" ], "description": "Get a ModelDeployment", "parameters": [ { "$ref": "../../../../../../common-types/resource-management/v6/types.json#/parameters/ApiVersionParameter" }, { "$ref": "../../../../../../common-types/resource-management/v6/types.json#/parameters/SubscriptionIdParameter" }, { "$ref": "../../../../../../common-types/resource-management/v6/types.json#/parameters/ResourceGroupNameParameter" }, { "name": "aiManagerName", "in": "path", "description": "The name of the AI Manager resource.", "required": true, "type": "string", "minLength": 1, "maxLength": 63, "pattern": "^[a-z0-9]([-a-z0-9]*[a-z0-9])?$" }, { "name": "namespaceName", "in": "path", "description": "The name of the AI Manager namespace resource.", "required": true, "type": "string", "minLength": 1, "maxLength": 63, "pattern": "^[a-z0-9]([-a-z0-9]*[a-z0-9])?$" }, { "name": "modelDeploymentName", "in": "path", "description": "The name of the model deployment resource.", "required": true, "type": "string", "minLength": 1, "maxLength": 63, "pattern": "^[a-z0-9]([-a-z0-9]*[a-z0-9])?$" } ], "responses": { "200": { "description": "Azure operation completed successfully.", "schema": { "$ref": "#/definitions/ModelDeployment" } }, "default": { "description": "An unexpected error response.", "schema": { "$ref": "../../../../../../common-types/resource-management/v6/types.json#/definitions/ErrorResponse" } } }, "x-ms-examples": { "ModelDeployments_Get_MaximumSet": { "$ref": "./examples/ModelDeployments_Get.json" } } }, "put": { "operationId": "ModelDeployments_CreateOrUpdate", "tags": [ "ModelDeployments" ], "description": "Create a ModelDeployment", "parameters": [ { "$ref": "../../../../../../common-types/resource-management/v6/types.json#/parameters/ApiVersionParameter" }, { "$ref": "../../../../../../common-types/resource-management/v6/types.json#/parameters/SubscriptionIdParameter" }, { "$ref": "../../../../../../common-types/resource-management/v6/types.json#/parameters/ResourceGroupNameParameter" }, { "name": "If-Match", "in": "header", "description": "The request should only proceed if an entity matches this string.", "required": false, "type": "string", "x-ms-client-name": "ifMatch" }, { "name": "If-None-Match", "in": "header", "description": "The request should only proceed if no entity matches this string.", "required": false, "type": "string", "x-ms-client-name": "ifNoneMatch" }, { "name": "aiManagerName", "in": "path", "description": "The name of the AI Manager resource.", "required": true, "type": "string", "minLength": 1, "maxLength": 63, "pattern": "^[a-z0-9]([-a-z0-9]*[a-z0-9])?$" }, { "name": "namespaceName", "in": "path", "description": "The name of the AI Manager namespace resource.", "required": true, "type": "string", "minLength": 1, "maxLength": 63, "pattern": "^[a-z0-9]([-a-z0-9]*[a-z0-9])?$" }, { "name": "modelDeploymentName", "in": "path", "description": "The name of the model deployment resource.", "required": true, "type": "string", "minLength": 1, "maxLength": 63, "pattern": "^[a-z0-9]([-a-z0-9]*[a-z0-9])?$" }, { "name": "resource", "in": "body", "description": "Resource create parameters.", "required": true, "schema": { "$ref": "#/definitions/ModelDeployment" } } ], "responses": { "200": { "description": "Resource 'ModelDeployment' update operation succeeded", "schema": { "$ref": "#/definitions/ModelDeployment" } }, "201": { "description": "Resource 'ModelDeployment' create operation succeeded", "schema": { "$ref": "#/definitions/ModelDeployment" }, "headers": { "Azure-AsyncOperation": { "type": "string", "description": "A link to the status monitor" }, "Retry-After": { "type": "integer", "format": "int32", "description": "The Retry-After header can indicate how long the client should wait before polling the operation status." } } }, "default": { "description": "An unexpected error response.", "schema": { "$ref": "../../../../../../common-types/resource-management/v6/types.json#/definitions/ErrorResponse" } } }, "x-ms-examples": { "ModelDeployments_CreateOrUpdate_MaximumSet": { "$ref": "./examples/ModelDeployments_CreateOrUpdate.json" } }, "x-ms-long-running-operation-options": { "final-state-via": "azure-async-operation" }, "x-ms-long-running-operation": true }, "patch": { "operationId": "ModelDeployments_Update", "tags": [ "ModelDeployments" ], "description": "Update a ModelDeployment", "parameters": [ { "$ref": "../../../../../../common-types/resource-management/v6/types.json#/parameters/ApiVersionParameter" }, { "$ref": "../../../../../../common-types/resource-management/v6/types.json#/parameters/SubscriptionIdParameter" }, { "$ref": "../../../../../../common-types/resource-management/v6/types.json#/parameters/ResourceGroupNameParameter" }, { "name": "If-Match", "in": "header", "description": "The request should only proceed if an entity matches this string.", "required": false, "type": "string", "x-ms-client-name": "ifMatch" }, { "name": "aiManagerName", "in": "path", "description": "The name of the AI Manager resource.", "required": true, "type": "string", "minLength": 1, "maxLength": 63, "pattern": "^[a-z0-9]([-a-z0-9]*[a-z0-9])?$" }, { "name": "namespaceName", "in": "path", "description": "The name of the AI Manager namespace resource.", "required": true, "type": "string", "minLength": 1, "maxLength": 63, "pattern": "^[a-z0-9]([-a-z0-9]*[a-z0-9])?$" }, { "name": "modelDeploymentName", "in": "path", "description": "The name of the model deployment resource.", "required": true, "type": "string", "minLength": 1, "maxLength": 63, "pattern": "^[a-z0-9]([-a-z0-9]*[a-z0-9])?$" }, { "name": "properties", "in": "body", "description": "The resource properties to be updated.", "required": true, "schema": { "$ref": "#/definitions/ModelDeploymentPatch" } } ], "responses": { "200": { "description": "Azure operation completed successfully.", "schema": { "$ref": "#/definitions/ModelDeployment" } }, "default": { "description": "An unexpected error response.", "schema": { "$ref": "../../../../../../common-types/resource-management/v6/types.json#/definitions/ErrorResponse" } } }, "x-ms-examples": { "ModelDeployments_Update_MaximumSet": { "$ref": "./examples/ModelDeployments_Update.json" } } }, "delete": { "operationId": "ModelDeployments_Delete", "tags": [ "ModelDeployments" ], "description": "Delete a ModelDeployment", "parameters": [ { "$ref": "../../../../../../common-types/resource-management/v6/types.json#/parameters/ApiVersionParameter" }, { "$ref": "../../../../../../common-types/resource-management/v6/types.json#/parameters/SubscriptionIdParameter" }, { "$ref": "../../../../../../common-types/resource-management/v6/types.json#/parameters/ResourceGroupNameParameter" }, { "name": "If-Match", "in": "header", "description": "The request should only proceed if an entity matches this string.", "required": false, "type": "string", "x-ms-client-name": "ifMatch" }, { "name": "aiManagerName", "in": "path", "description": "The name of the AI Manager resource.", "required": true, "type": "string", "minLength": 1, "maxLength": 63, "pattern": "^[a-z0-9]([-a-z0-9]*[a-z0-9])?$" }, { "name": "namespaceName", "in": "path", "description": "The name of the AI Manager namespace resource.", "required": true, "type": "string", "minLength": 1, "maxLength": 63, "pattern": "^[a-z0-9]([-a-z0-9]*[a-z0-9])?$" }, { "name": "modelDeploymentName", "in": "path", "description": "The name of the model deployment resource.", "required": true, "type": "string", "minLength": 1, "maxLength": 63, "pattern": "^[a-z0-9]([-a-z0-9]*[a-z0-9])?$" } ], "responses": { "202": { "description": "Resource deletion accepted.", "headers": { "Location": { "type": "string", "description": "The Location header contains the URL where the status of the long running operation can be checked." }, "Retry-After": { "type": "integer", "format": "int32", "description": "The Retry-After header can indicate how long the client should wait before polling the operation status." } } }, "204": { "description": "Resource does not exist." }, "default": { "description": "An unexpected error response.", "schema": { "$ref": "../../../../../../common-types/resource-management/v6/types.json#/definitions/ErrorResponse" } } }, "x-ms-examples": { "ModelDeployments_Delete_MaximumSet": { "$ref": "./examples/ModelDeployments_Delete.json" } }, "x-ms-long-running-operation-options": { "final-state-via": "location" }, "x-ms-long-running-operation": true } }, "/subscriptions/{subscriptionId}/resourceGroups/{resourceGroupName}/providers/Microsoft.ContainerService/aiModels": { "get": { "operationId": "AIModels_ListByResourceGroup", "tags": [ "AIModels" ], "description": "List AIModel resources by resource group", "parameters": [ { "$ref": "../../../../../../common-types/resource-management/v6/types.json#/parameters/ApiVersionParameter" }, { "$ref": "../../../../../../common-types/resource-management/v6/types.json#/parameters/SubscriptionIdParameter" }, { "$ref": "../../../../../../common-types/resource-management/v6/types.json#/parameters/ResourceGroupNameParameter" } ], "responses": { "200": { "description": "Azure operation completed successfully.", "schema": { "$ref": "#/definitions/AIModelListResult" } }, "default": { "description": "An unexpected error response.", "schema": { "$ref": "../../../../../../common-types/resource-management/v6/types.json#/definitions/ErrorResponse" } } }, "x-ms-examples": { "AIModels_ListByResourceGroup_MaximumSet": { "$ref": "./examples/AIModels_ListByResourceGroup.json" } }, "x-ms-pageable": { "nextLinkName": "nextLink" } } }, "/subscriptions/{subscriptionId}/resourceGroups/{resourceGroupName}/providers/Microsoft.ContainerService/aiModels/{aiModelName}": { "get": { "operationId": "AIModels_Get", "tags": [ "AIModels" ], "description": "Get a AIModel", "parameters": [ { "$ref": "../../../../../../common-types/resource-management/v6/types.json#/parameters/ApiVersionParameter" }, { "$ref": "../../../../../../common-types/resource-management/v6/types.json#/parameters/SubscriptionIdParameter" }, { "$ref": "../../../../../../common-types/resource-management/v6/types.json#/parameters/ResourceGroupNameParameter" }, { "name": "aiModelName", "in": "path", "description": "The name of the AI model resource. Matches the upstream model id (e.g. \"microsoft/Phi-4-mini-instruct\"). The `/` character must be URL-encoded as `%2F` on the wire.", "required": true, "type": "string", "minLength": 1, "maxLength": 128, "pattern": "^[A-Za-z0-9][A-Za-z0-9._-]*(/[A-Za-z0-9][A-Za-z0-9._-]*)?$" } ], "responses": { "200": { "description": "Azure operation completed successfully.", "schema": { "$ref": "#/definitions/AIModel" } }, "default": { "description": "An unexpected error response.", "schema": { "$ref": "../../../../../../common-types/resource-management/v6/types.json#/definitions/ErrorResponse" } } }, "x-ms-examples": { "AIModels_Get_MaximumSet": { "$ref": "./examples/AIModels_Get.json" } } } }, "/subscriptions/{subscriptionId}/resourceGroups/{resourceGroupName}/providers/Microsoft.ContainerService/aiModels/{aiModelName}/calculateCost": { "post": { "operationId": "AIModels_CalculateCost", "tags": [ "AIModels" ], "description": "Calculates projected hourly compute cost for deploying this model on the GPU SKUs available to the caller. No Kubernetes or Azure resources are provisioned.", "parameters": [ { "$ref": "../../../../../../common-types/resource-management/v6/types.json#/parameters/ApiVersionParameter" }, { "$ref": "../../../../../../common-types/resource-management/v6/types.json#/parameters/SubscriptionIdParameter" }, { "$ref": "../../../../../../common-types/resource-management/v6/types.json#/parameters/ResourceGroupNameParameter" }, { "name": "aiModelName", "in": "path", "description": "The name of the AI model resource. Matches the upstream model id (e.g. \"microsoft/Phi-4-mini-instruct\"). The `/` character must be URL-encoded as `%2F` on the wire.", "required": true, "type": "string", "minLength": 1, "maxLength": 128, "pattern": "^[A-Za-z0-9][A-Za-z0-9._-]*(/[A-Za-z0-9][A-Za-z0-9._-]*)?$" }, { "name": "body", "in": "body", "description": "The content of the action request", "required": true, "schema": { "$ref": "#/definitions/CalculateCostRequest" } } ], "responses": { "200": { "description": "Azure operation completed successfully.", "schema": { "$ref": "#/definitions/CalculateCostResponse" } }, "default": { "description": "An unexpected error response.", "schema": { "$ref": "../../../../../../common-types/resource-management/v6/types.json#/definitions/ErrorResponse" } } }, "x-ms-examples": { "AIModels_CalculateCost_MaximumSet": { "$ref": "./examples/AIModels_CalculateCost.json" } } } } }, "definitions": { "AIManager": { "type": "object", "description": "The AI Manager resource. For more information, see https://aka.ms/aks/aimanager.", "properties": { "properties": { "$ref": "#/definitions/AIManagerProperties", "description": "The resource-specific properties for this resource." }, "eTag": { "type": "string", "description": "If eTag is provided in the response body, it may also be provided as a header per the normal etag convention. Entity tags are used for comparing two or more entities from the same requested resource. HTTP/1.1 uses entity tags in the etag (section 14.19), If-Match (section 14.24), If-None-Match (section 14.26), and If-Range (section 14.27) header fields.", "readOnly": true }, "identity": { "$ref": "../../../../../../common-types/resource-management/v6/managedidentity.json#/definitions/ManagedServiceIdentity", "description": "The managed service identities assigned to this resource." } }, "allOf": [ { "$ref": "../../../../../../common-types/resource-management/v6/types.json#/definitions/TrackedResource" } ] }, "AIManagerListResult": { "type": "object", "description": "The response of a AIManager list operation.", "properties": { "value": { "type": "array", "description": "The AIManager items on this page", "items": { "$ref": "#/definitions/AIManager" } }, "nextLink": { "type": "string", "format": "uri", "description": "The link to the next page of items" } }, "required": [ "value" ] }, "AIManagerNamespace": { "type": "object", "description": "The AI Manager namespace resource.", "properties": { "properties": { "$ref": "#/definitions/AIManagerNamespaceProperties", "description": "The resource-specific properties for this resource." }, "eTag": { "type": "string", "description": "If eTag is provided in the response body, it may also be provided as a header per the normal etag convention. Entity tags are used for comparing two or more entities from the same requested resource. HTTP/1.1 uses entity tags in the etag (section 14.19), If-Match (section 14.24), If-None-Match (section 14.26), and If-Range (section 14.27) header fields.", "readOnly": true } }, "allOf": [ { "$ref": "../../../../../../common-types/resource-management/v6/types.json#/definitions/ProxyResource" } ] }, "AIManagerNamespaceListResult": { "type": "object", "description": "The response of a AIManagerNamespace list operation.", "properties": { "value": { "type": "array", "description": "The AIManagerNamespace items on this page", "items": { "$ref": "#/definitions/AIManagerNamespace" } }, "nextLink": { "type": "string", "format": "uri", "description": "The link to the next page of items" } }, "required": [ "value" ] }, "AIManagerNamespaceProperties": { "type": "object", "description": "AI Manager namespace properties.", "properties": { "provisioningState": { "$ref": "#/definitions/AIManagerNamespaceProvisioningState", "description": "The status of the last operation.", "readOnly": true }, "labels": { "type": "object", "description": "Labels applied to the Kubernetes namespace.", "additionalProperties": { "type": "string" } }, "annotations": { "type": "object", "description": "Annotations applied to the Kubernetes namespace.", "additionalProperties": { "type": "string" } } } }, "AIManagerNamespaceProvisioningState": { "type": "string", "description": "The provisioning state of the AI Manager namespace resource.", "enum": [ "Succeeded", "Failed", "Canceled", "Creating", "Updating", "Deleting" ], "x-ms-enum": { "name": "AIManagerNamespaceProvisioningState", "modelAsString": true, "values": [ { "name": "Succeeded", "value": "Succeeded", "description": "Resource has been created." }, { "name": "Failed", "value": "Failed", "description": "Resource creation failed." }, { "name": "Canceled", "value": "Canceled", "description": "Resource creation was canceled." }, { "name": "Creating", "value": "Creating", "description": "The provisioning state of a namespace being created." }, { "name": "Updating", "value": "Updating", "description": "The provisioning state of a namespace being updated." }, { "name": "Deleting", "value": "Deleting", "description": "The provisioning state of a namespace being deleted." } ] }, "readOnly": true }, "AIManagerPatch": { "type": "object", "description": "The AI Manager resource patch model.", "properties": { "tags": { "type": "object", "description": "Resource tags.", "additionalProperties": { "type": "string" } }, "identity": { "$ref": "../../../../../../common-types/resource-management/v6/managedidentity.json#/definitions/ManagedServiceIdentity", "description": "The managed service identities assigned to this resource." } } }, "AIManagerProperties": { "type": "object", "description": "AI Manager properties.", "properties": { "provisioningState": { "$ref": "#/definitions/AIManagerProvisioningState", "description": "The status of the last operation.", "readOnly": true }, "deletePolicy": { "$ref": "#/definitions/DeletePolicy", "description": "Delete options of the AI Manager. Defaults to `Delete` if not specified." }, "managedResourceGroupName": { "type": "string", "description": "The name of the managed resource group created by the AI Manager to hold underlying infrastructure resources.", "readOnly": true } } }, "AIManagerProvisioningState": { "type": "string", "description": "The provisioning state of the AI Manager resource.", "enum": [ "Succeeded", "Failed", "Canceled", "Creating", "Updating", "Deleting" ], "x-ms-enum": { "name": "AIManagerProvisioningState", "modelAsString": true, "values": [ { "name": "Succeeded", "value": "Succeeded", "description": "Resource has been created." }, { "name": "Failed", "value": "Failed", "description": "Resource creation failed." }, { "name": "Canceled", "value": "Canceled", "description": "Resource creation was canceled." }, { "name": "Creating", "value": "Creating", "description": "Resource is being created." }, { "name": "Updating", "value": "Updating", "description": "Resource is updating." }, { "name": "Deleting", "value": "Deleting", "description": "Resource is deleting." } ] }, "readOnly": true }, "AIModel": { "type": "object", "description": "An AI model exposed by Microsoft.ContainerService. Platform-maintained and auto-provisioned by the resource provider; can be referenced by `ModelDeployment` resources.", "properties": { "properties": { "$ref": "#/definitions/AIModelProperties", "description": "The resource-specific properties for this resource." } }, "allOf": [ { "$ref": "../../../../../../common-types/resource-management/v6/types.json#/definitions/TrackedResource" } ] }, "AIModelListResult": { "type": "object", "description": "The response of a AIModel list operation.", "properties": { "value": { "type": "array", "description": "The AIModel items on this page", "items": { "$ref": "#/definitions/AIModel" } }, "nextLink": { "type": "string", "format": "uri", "description": "The link to the next page of items" } }, "required": [ "value" ] }, "AIModelProperties": { "type": "object", "description": "AI model properties.", "properties": { "description": { "type": "string", "description": "An optional, free-form description of the model." }, "resolvedSpec": { "$ref": "#/definitions/ResolvedModelSpec", "description": "Platform-resolved specification of the model.", "readOnly": true } } }, "AIModelResourceId": { "type": "string", "format": "arm-id", "description": "The ARM resource id of an AIModel.", "x-ms-arm-id-details": { "allowedResources": [ { "type": "Microsoft.ContainerService/aiModels" } ] } }, "AutoscalingProfile": { "type": "object", "description": "Autoscaling configuration for a model deployment.", "properties": { "enabled": { "type": "boolean", "description": "Whether autoscaling is enabled for this deployment.", "default": false }, "minReplicas": { "type": "integer", "format": "int32", "description": "The minimum number of replicas when autoscaling is enabled.", "default": 1, "minimum": 1 }, "maxReplicas": { "type": "integer", "format": "int32", "description": "The maximum number of replicas when autoscaling is enabled. If not specified, the service derives a default from the subscription GPU quota.", "minimum": 1 } } }, "Azure.ResourceManager.ResourceProvisioningState": { "type": "string", "description": "The provisioning state of a resource type.", "enum": [ "Succeeded", "Failed", "Canceled" ], "x-ms-enum": { "name": "ResourceProvisioningState", "modelAsString": true, "values": [ { "name": "Succeeded", "value": "Succeeded", "description": "Resource has been created." }, { "name": "Failed", "value": "Failed", "description": "Resource creation failed." }, { "name": "Canceled", "value": "Canceled", "description": "Resource creation was canceled." } ] }, "readOnly": true }, "CalculateCostPlan": { "type": "object", "description": "A GPU SKU pricing plan returned by the `calculateCost` action.", "properties": { "vmSize": { "type": "string", "description": "Azure VM SKU, e.g. \"Standard_ND96isr_H100_v5\". Matches the value accepted by `ModelDeploymentProperties.vmSize`." }, "quantization": { "type": "string", "description": "Resolved quantization on this SKU." }, "vmsPerReplica": { "type": "integer", "format": "int32", "description": "Number of cluster nodes required to host one replica on this SKU." }, "vmHourlyPrice": { "type": "number", "format": "double", "description": "On-demand hourly price for a single node of this SKU, in `currency`." }, "totalHourlyPrice": { "type": "number", "format": "double", "description": "Projected hourly cost for `replicas * vmsPerReplica` nodes, in `currency`. Omitted when the plan is infeasible." }, "priceAsOf": { "type": "string", "format": "date-time", "description": "UTC timestamp of the price snapshot used for this plan. Overrides any response-level snapshot if present." }, "feasible": { "type": "boolean", "description": "Whether the caller can actually deploy this plan today (region availability, GPU quota, model fit, ...)." }, "infeasibleCode": { "$ref": "#/definitions/InfeasibleCode", "description": "Machine-readable reason when `feasible=false`. Omitted when `feasible=true`." }, "infeasibleMessage": { "type": "string", "description": "Human-readable message accompanying `infeasibleCode`. Omitted when `feasible=true`." } }, "required": [ "vmSize", "vmsPerReplica", "vmHourlyPrice", "feasible" ] }, "CalculateCostRequest": { "type": "object", "description": "Request body for the AI model `calculateCost` action.", "properties": { "replicas": { "type": "integer", "format": "int32", "description": "Number of replicas to price.", "default": 1, "minimum": 1 } } }, "CalculateCostResponse": { "type": "object", "description": "Response body for the AI model `calculateCost` action.", "properties": { "replicas": { "type": "integer", "format": "int32", "description": "Echoes the effective replica count used for totals." }, "currency": { "type": "string", "description": "ISO 4217 currency code, e.g. \"USD\"." }, "plans": { "type": "array", "description": "Ranked list of GPU SKU pricing plans. Feasible plans first, ordered by `totalHourlyPrice` ascending; infeasible plans last.", "items": { "$ref": "#/definitions/CalculateCostPlan" }, "x-ms-identifiers": [ "vmSize" ] } }, "required": [ "replicas", "currency", "plans" ] }, "CredentialResult": { "type": "object", "description": "The credential result response.", "properties": { "name": { "type": "string", "description": "The name of the credential.", "readOnly": true }, "value": { "type": "string", "format": "password", "description": "Base64-encoded Kubernetes configuration file.", "readOnly": true, "x-ms-secret": true } } }, "CredentialResults": { "type": "object", "description": "The list credential result response.", "properties": { "kubeconfigs": { "type": "array", "description": "Array of credential results.", "items": { "$ref": "#/definitions/CredentialResult" }, "readOnly": true, "x-ms-identifiers": [] } } }, "CredentialSourceType": { "type": "string", "description": "The credential source type.", "enum": [ "Inline" ], "x-ms-enum": { "name": "CredentialSourceType", "modelAsString": true, "values": [ { "name": "Inline", "value": "Inline", "description": "A credential provided inline as a secret value in the request payload." } ] } }, "CredentialValue": { "type": "object", "description": "A credential value, discriminated by `type`.", "properties": { "type": { "$ref": "#/definitions/CredentialSourceType", "description": "The credential source type." } }, "discriminator": "type", "required": [ "type" ] }, "DeletePolicy": { "type": "string", "description": "Delete options of the AI Manager.", "enum": [ "Keep", "Delete" ], "x-ms-enum": { "name": "DeletePolicy", "modelAsString": true, "values": [ { "name": "Keep", "value": "Keep", "description": "Keep the underlying cluster resources even if the AIManager resource is deleted." }, { "name": "Delete", "value": "Delete", "description": "Delete both the underlying cluster resources and the AIManager resource together." } ] } }, "InfeasibleCode": { "type": "string", "description": "The reason a `CalculateCostPlan` is not deployable.", "enum": [ "InsufficientQuota", "RegionUnavailable", "InefficientDeployment" ], "x-ms-enum": { "name": "InfeasibleCode", "modelAsString": true, "values": [ { "name": "InsufficientQuota", "value": "InsufficientQuota", "description": "The caller's subscription does not have enough GPU quota in the target region to deploy this plan." }, { "name": "RegionUnavailable", "value": "RegionUnavailable", "description": "The VM SKU is not available in the target region." }, { "name": "InefficientDeployment", "value": "InefficientDeployment", "description": "The deployment can start successfully on this SKU, but its measured runtime performance falls below the acceptable threshold for serving this model." } ] } }, "InlineCredential": { "type": "object", "description": "A credential provided inline.", "properties": { "value": { "type": "string", "format": "password", "description": "The access token, password, or other secret value.", "x-ms-secret": true } }, "required": [ "value" ], "allOf": [ { "$ref": "#/definitions/CredentialValue" } ], "x-ms-discriminator-value": "Inline" }, "ModelDeployment": { "type": "object", "description": "A running deployment of a model in an AI Manager namespace.", "properties": { "properties": { "$ref": "#/definitions/ModelDeploymentProperties", "description": "The resource-specific properties for this resource." }, "eTag": { "type": "string", "description": "If eTag is provided in the response body, it may also be provided as a header per the normal etag convention. Entity tags are used for comparing two or more entities from the same requested resource. HTTP/1.1 uses entity tags in the etag (section 14.19), If-Match (section 14.24), If-None-Match (section 14.26), and If-Range (section 14.27) header fields.", "readOnly": true } }, "allOf": [ { "$ref": "../../../../../../common-types/resource-management/v6/types.json#/definitions/ProxyResource" } ] }, "ModelDeploymentListResult": { "type": "object", "description": "The response of a ModelDeployment list operation.", "properties": { "value": { "type": "array", "description": "The ModelDeployment items on this page", "items": { "$ref": "#/definitions/ModelDeployment" } }, "nextLink": { "type": "string", "format": "uri", "description": "The link to the next page of items" } }, "required": [ "value" ] }, "ModelDeploymentOverrides": { "type": "object", "description": "User overrides for a model deployment.", "properties": { "values": { "type": "object", "description": "Free-form override key/value pairs. Recognized keys are documented per release.", "additionalProperties": { "type": "string" } } } }, "ModelDeploymentPatch": { "type": "object", "description": "The model deployment resource patch model.", "properties": { "properties": { "$ref": "#/definitions/ModelDeploymentPatchProperties", "description": "Mutable properties of the model deployment." } } }, "ModelDeploymentPatchProperties": { "type": "object", "description": "Mutable properties of a model deployment.", "properties": { "performanceMode": { "$ref": "#/definitions/ModelDeploymentPerformanceMode", "description": "Runtime performance mode." }, "replicas": { "type": "integer", "format": "int32", "description": "Desired replica count. Ignored when `autoscaling.enabled` is true.", "minimum": 1 }, "autoscaling": { "$ref": "#/definitions/AutoscalingProfile", "description": "The autoscaling configuration for the deployment." }, "overrides": { "$ref": "#/definitions/ModelDeploymentOverrides", "description": "User overrides layered on top of profile resolution." } } }, "ModelDeploymentPerformanceMode": { "type": "string", "description": "The runtime performance mode of a model deployment.", "enum": [ "Balanced", "Latency", "Throughput" ], "x-ms-enum": { "name": "ModelDeploymentPerformanceMode", "modelAsString": true, "values": [ { "name": "Balanced", "value": "Balanced", "description": "A balanced trade-off between latency and throughput (default)." }, { "name": "Latency", "value": "Latency", "description": "Optimize for low request latency." }, { "name": "Throughput", "value": "Throughput", "description": "Optimize for high aggregate throughput." } ] } }, "ModelDeploymentProperties": { "type": "object", "description": "Model deployment properties.", "properties": { "provisioningState": { "$ref": "#/definitions/ModelDeploymentProvisioningState", "description": "The status of the last reconciliation.", "readOnly": true }, "modelResourceId": { "$ref": "#/definitions/AIModelResourceId", "description": "Full ARM resource id of the model to deploy. Phase 1 accepts an `AIModel` resource id only. Immutable after creation.", "x-ms-mutability": [ "read", "create" ] }, "modelSourceResourceId": { "$ref": "#/definitions/ModelSourceResourceId", "description": "Full ARM resource id of a `ModelSource` to use when pulling artifacts for this deployment. Immutable after creation.", "x-ms-mutability": [ "read", "create" ] }, "performanceMode": { "type": "string", "description": "Runtime performance mode. Selects a default engine/quantization combination; use `overrides` to pin values.", "default": "Balanced", "enum": [ "Balanced", "Latency", "Throughput" ], "x-ms-enum": { "name": "ModelDeploymentPerformanceMode", "modelAsString": true, "values": [ { "name": "Balanced", "value": "Balanced", "description": "A balanced trade-off between latency and throughput (default)." }, { "name": "Latency", "value": "Latency", "description": "Optimize for low request latency." }, { "name": "Throughput", "value": "Throughput", "description": "Optimize for high aggregate throughput." } ] } }, "vmSize": { "type": "string", "description": "Azure VM SKU used to host the deployment, e.g. \"Standard_NC96ads_A100_v4\". Immutable after creation.", "x-ms-mutability": [ "read", "create" ] }, "replicas": { "type": "integer", "format": "int32", "description": "Desired replica count. Ignored when `autoscaling.enabled` is true.", "default": 1, "minimum": 1 }, "autoscaling": { "$ref": "#/definitions/AutoscalingProfile", "description": "The autoscaling configuration for the deployment." }, "overrides": { "$ref": "#/definitions/ModelDeploymentOverrides", "description": "User overrides layered on top of profile resolution. Replace semantics on PATCH: the entire object is replaced." }, "status": { "$ref": "#/definitions/ModelDeploymentStatus", "description": "Runtime status, populated once reconciliation begins.", "readOnly": true } }, "required": [ "modelResourceId", "vmSize" ] }, "ModelDeploymentProvisioningState": { "type": "string", "description": "The provisioning state of a model deployment resource.", "enum": [ "Succeeded", "Failed", "Canceled", "Creating", "Updating", "Deleting" ], "x-ms-enum": { "name": "ModelDeploymentProvisioningState", "modelAsString": true, "values": [ { "name": "Succeeded", "value": "Succeeded", "description": "Resource has been created." }, { "name": "Failed", "value": "Failed", "description": "Resource creation failed." }, { "name": "Canceled", "value": "Canceled", "description": "Resource creation was canceled." }, { "name": "Creating", "value": "Creating", "description": "Resource is being created." }, { "name": "Updating", "value": "Updating", "description": "Resource is updating." }, { "name": "Deleting", "value": "Deleting", "description": "Resource is deleting." } ] }, "readOnly": true }, "ModelDeploymentStatus": { "type": "object", "description": "The runtime status of a model deployment. All fields are read-only and populated once reconciliation has started.", "properties": { "endpoint": { "type": "string", "format": "uri", "description": "The inference endpoint URL exposed by the deployment, once ready.", "readOnly": true }, "engine": { "type": "string", "description": "The inference engine used to serve the model, e.g. \"vllm\".", "readOnly": true }, "engineVersion": { "type": "string", "description": "The version of the inference engine, e.g. \"0.17\".", "readOnly": true }, "maxModelLen": { "type": "integer", "format": "int32", "description": "The maximum model context length, in tokens, configured for this deployment.", "readOnly": true }, "quantization": { "type": "string", "description": "The quantization level applied to the model weights, e.g. \"fp16\", \"awq-int4\".", "readOnly": true }, "desiredReplicas": { "type": "integer", "format": "int32", "description": "The desired replica count reported by the controller. Equals `properties.replicas` when autoscaler is disabled; current target replica count otherwise.", "readOnly": true }, "peakTokensPerMinute": { "type": "integer", "format": "int32", "description": "The peak tokens per minute measured by live stress test.", "readOnly": true }, "estimatedProvisionTimeSeconds": { "type": "integer", "format": "int32", "description": "Estimated total time, in seconds, for the deployment to become ready end-to-end (GPU node provisioning, image/weight pull, engine warm-up).", "readOnly": true } } }, "ModelSource": { "type": "object", "description": "A model source registered with an AI Manager. Describes an external model registry (e.g. Hugging Face) and the credentials the platform uses to pull artifacts from it.", "properties": { "properties": { "$ref": "#/definitions/ModelSourceProperties", "description": "The resource-specific properties for this resource." }, "eTag": { "type": "string", "description": "If eTag is provided in the response body, it may also be provided as a header per the normal etag convention. Entity tags are used for comparing two or more entities from the same requested resource. HTTP/1.1 uses entity tags in the etag (section 14.19), If-Match (section 14.24), If-None-Match (section 14.26), and If-Range (section 14.27) header fields.", "readOnly": true } }, "allOf": [ { "$ref": "../../../../../../common-types/resource-management/v6/types.json#/definitions/ProxyResource" } ] }, "ModelSourceListResult": { "type": "object", "description": "The response of a ModelSource list operation.", "properties": { "value": { "type": "array", "description": "The ModelSource items on this page", "items": { "$ref": "#/definitions/ModelSource" } }, "nextLink": { "type": "string", "format": "uri", "description": "The link to the next page of items" } }, "required": [ "value" ] }, "ModelSourcePatch": { "type": "object", "description": "The model source resource patch model.", "properties": { "properties": { "$ref": "#/definitions/ModelSourcePatchProperties", "description": "Mutable properties of the model source." } } }, "ModelSourcePatchProperties": { "type": "object", "description": "Mutable properties of a model source.", "properties": { "description": { "type": "string", "description": "An optional, free-form description of the source." }, "credential": { "$ref": "#/definitions/CredentialValue", "description": "Credential the platform uses to authenticate to the source." } } }, "ModelSourceProperties": { "type": "object", "description": "Model source properties.", "properties": { "provisioningState": { "$ref": "#/definitions/Azure.ResourceManager.ResourceProvisioningState", "description": "The status of the last operation.", "readOnly": true }, "sourceType": { "$ref": "#/definitions/ModelSourceType", "description": "Model source type. Constrains the legal authentication kinds. Immutable after creation.", "x-ms-mutability": [ "read", "create" ] }, "description": { "type": "string", "description": "An optional, free-form description of the source." }, "credential": { "$ref": "#/definitions/CredentialValue", "description": "Credential the platform uses to authenticate to the source. Optional for public sources (e.g. ungated Hugging Face models)." } }, "required": [ "sourceType" ] }, "ModelSourceResourceId": { "type": "string", "format": "arm-id", "description": "The ARM resource id of a ModelSource.", "x-ms-arm-id-details": { "allowedResources": [ { "type": "Microsoft.ContainerService/aiManagers/modelSources" } ] } }, "ModelSourceType": { "type": "string", "description": "The type of a model source.", "enum": [ "HuggingFace" ], "x-ms-enum": { "name": "ModelSourceType", "modelAsString": true, "values": [ { "name": "HuggingFace", "value": "HuggingFace", "description": "A Hugging Face model registry." } ] } }, "NamespaceAccessInfo": { "type": "object", "description": "Access information for an AI Manager namespace, including the OpenAI-compatible gateway endpoint and the API keys used to authenticate against it.", "properties": { "endpoint": { "type": "string", "format": "uri", "description": "OpenAI-compatible base URL, e.g. \"https://team-alpha...aksapp.io/v1\"." }, "primaryKey": { "type": "string", "format": "password", "description": "Primary API key. Send as `Authorization: Bearer ` (or `api-key: `, depending on client). Treat as secret; do not log or persist in plaintext.", "x-ms-secret": true }, "secondaryKey": { "type": "string", "format": "password", "description": "Secondary API key, for zero-downtime rotation. Same usage as `primaryKey`.", "x-ms-secret": true }, "lastRotatedAt": { "type": "string", "format": "date-time", "description": "UTC time the keys were last rotated. Clients can use this to detect rotation and refresh cached credentials." } }, "required": [ "endpoint", "primaryKey", "secondaryKey" ] }, "ResolvedModelSpec": { "type": "object", "description": "The platform-resolved specification of a model. All fields are read-only.", "properties": { "license": { "type": "string", "description": "The license of the model, when known.", "readOnly": true }, "gated": { "type": "boolean", "description": "Whether the model is gated and requires explicit access approval.", "readOnly": true }, "maxContextLength": { "type": "integer", "format": "int32", "description": "The maximum context length supported by the model, in tokens.", "readOnly": true } }, "required": [ "maxContextLength" ] } }, "parameters": {} }