Adding support for Azure OpenAI Deployment Types (Global Standard, Standard, Provisioned)

.azdo/pipelines/azure-dev.yml

@@ -66,11 +66,18 @@ steps:
       AZURE_OPENAI_CHATGPT_DEPLOYMENT: $(AZURE_OPENAI_CHATGPT_DEPLOYMENT)
       AZURE_OPENAI_CHATGPT_DEPLOYMENT_CAPACITY: $(AZURE_OPENAI_CHATGPT_DEPLOYMENT_CAPACITY)
       AZURE_OPENAI_CHATGPT_DEPLOYMENT_VERSION: $(AZURE_OPENAI_CHATGPT_DEPLOYMENT_VERSION)
+      AZURE_OPENAI_CHATGPT_DEPLOYMENT_SKU: $(AZURE_OPENAI_CHATGPT_DEPLOYMENT_SKU)
       AZURE_OPENAI_EMB_MODEL_NAME: $(AZURE_OPENAI_EMB_MODEL_NAME)
       AZURE_OPENAI_EMB_DEPLOYMENT: $(AZURE_OPENAI_EMB_DEPLOYMENT)
       AZURE_OPENAI_EMB_DEPLOYMENT_CAPACITY: $(AZURE_OPENAI_EMB_DEPLOYMENT_CAPACITY)
       AZURE_OPENAI_EMB_DEPLOYMENT_VERSION: $(AZURE_OPENAI_EMB_DEPLOYMENT_VERSION)
+      AZURE_OPENAI_EMB_DEPLOYMENT_SKU: $(AZURE_OPENAI_EMB_DEPLOYMENT_SKU)
       AZURE_OPENAI_EMB_DIMENSIONS: $(AZURE_OPENAI_EMB_DIMENSIONS)
+      AZURE_OPENAI_GPT4V_MODEL: $(AZURE_OPENAI_GPT4V_MODEL)
+      AZURE_OPENAI_GPT4V_DEPLOYMENT: $(AZURE_OPENAI_GPT4V_DEPLOYMENT)
+      AZURE_OPENAI_GPT4V_DEPLOYMENT_CAPACITY: $(AZURE_OPENAI_GPT4V_DEPLOYMENT_CAPACITY)
+      AZURE_OPENAI_GPT4V_DEPLOYMENT_VERSION: $(AZURE_OPENAI_GPT4V_DEPLOYMENT_VERSION)
+      AZURE_OPENAI_GPT4V_DEPLOYMENT_SKU: $(AZURE_OPENAI_GPT4V_DEPLOYMENT_SKU)
       OPENAI_HOST: $(OPENAI_HOST)
       OPENAI_API_KEY: $(OPENAI_API_KEY)
       OPENAI_ORGANIZATION: $(OPENAI_ORGANIZATION)

azure.yaml

@@ -64,11 +64,18 @@ pipeline:
       - AZURE_OPENAI_CHATGPT_DEPLOYMENT
       - AZURE_OPENAI_CHATGPT_DEPLOYMENT_CAPACITY
       - AZURE_OPENAI_CHATGPT_DEPLOYMENT_VERSION
+      - AZURE_OPENAI_CHATGPT_DEPLOYMENT_SKU
       - AZURE_OPENAI_EMB_MODEL_NAME
       - AZURE_OPENAI_EMB_DEPLOYMENT
       - AZURE_OPENAI_EMB_DEPLOYMENT_CAPACITY
       - AZURE_OPENAI_EMB_DEPLOYMENT_VERSION
+      - AZURE_OPENAI_EMB_DEPLOYMENT_SKU
       - AZURE_OPENAI_EMB_DIMENSIONS
+      - AZURE_OPENAI_GPT4V_MODEL
+      - AZURE_OPENAI_GPT4V_DEPLOYMENT
+      - AZURE_OPENAI_GPT4V_DEPLOYMENT_CAPACITY
+      - AZURE_OPENAI_GPT4V_DEPLOYMENT_VERSION
+      - AZURE_OPENAI_GPT4V_DEPLOYMENT_SKU
       - OPENAI_HOST
       - OPENAI_API_KEY
       - OPENAI_ORGANIZATION

docs/deploy_features.md

@@ -59,6 +59,12 @@ Execute the following commands inside your terminal:
     azd env set AZURE_OPENAI_CHATGPT_MODEL gpt-4o-mini
     ```
 
+1. To set the Azure OpenAI deployment SKU name, run this command with [the desired SKU name](https://learn.microsoft.com/azure/ai-services/openai/how-to/deployment-types#deployment-types).
+
+    ```bash
+    azd env set AZURE_OPENAI_CHATGPT_DEPLOYMENT_SKU GlobalStandard
+    ```
+
 1. To set the Azure OpenAI deployment capacity, run this command with the desired capacity.
 
     ```bash
@@ -97,6 +103,7 @@ Execute the following commands inside your terminal:
 > * `azd env set AZURE_OPENAI_CHATGPT_DEPLOYMENT chat` to set the name of your old GPT 3.5 deployment.
 > * `azd env set AZURE_OPENAI_CHATGPT_MODEL gpt-35-turbo` to set the name of your old GPT 3.5 model.
 > * `azd env set AZURE_OPENAI_CHATGPT_DEPLOYMENT_CAPACITY 30` to set the capacity of your old GPT 3.5 deployment.
+> * `azd env set AZURE_OPENAI_CHATGPT_DEPLOYMENT_SKU Standard` to set the Sku name back to Standard.
 > * `azd env set AZURE_OPENAI_CHATGPT_DEPLOYMENT_VERSION 0613` to set the version number of your old GPT 3.5.
 > * `azd up` to update the provisioned resources.
 >

infra/main.bicep

@@ -109,33 +109,46 @@ param computerVisionSkuName string // Set in main.parameters.json
 param chatGptModelName string = ''
 param chatGptDeploymentName string = ''
 param chatGptDeploymentVersion string = ''
+param chatGptDeploymentSkuName string = ''
 param chatGptDeploymentCapacity int = 0
+
 var chatGpt = {
   modelName: !empty(chatGptModelName)
     ? chatGptModelName
     : startsWith(openAiHost, 'azure') ? 'gpt-35-turbo' : 'gpt-3.5-turbo'
   deploymentName: !empty(chatGptDeploymentName) ? chatGptDeploymentName : 'chat'
   deploymentVersion: !empty(chatGptDeploymentVersion) ? chatGptDeploymentVersion : '0613'
+  deploymentSkuName: !empty(chatGptDeploymentSkuName) ? chatGptDeploymentSkuName : 'Standard'
   deploymentCapacity: chatGptDeploymentCapacity != 0 ? chatGptDeploymentCapacity : 30
 }
 
 param embeddingModelName string = ''
 param embeddingDeploymentName string = ''
 param embeddingDeploymentVersion string = ''
+param embeddingDeploymentSkuName string = ''
 param embeddingDeploymentCapacity int = 0
 param embeddingDimensions int = 0
 var embedding = {
   modelName: !empty(embeddingModelName) ? embeddingModelName : 'text-embedding-ada-002'
   deploymentName: !empty(embeddingDeploymentName) ? embeddingDeploymentName : 'embedding'
   deploymentVersion: !empty(embeddingDeploymentVersion) ? embeddingDeploymentVersion : '2'
+  deploymentSkuName: !empty(embeddingDeploymentSkuName) ? embeddingDeploymentSkuName : 'Standard'
   deploymentCapacity: embeddingDeploymentCapacity != 0 ? embeddingDeploymentCapacity : 30
   dimensions: embeddingDimensions != 0 ? embeddingDimensions : 1536
 }
 
-param gpt4vModelName string = 'gpt-4o'
-param gpt4vDeploymentName string = 'gpt-4o'
-param gpt4vModelVersion string = '2024-05-13'
-param gpt4vDeploymentCapacity int = 10
+param gpt4vModelName string = ''
+param gpt4vDeploymentName string = ''
+param gpt4vModelVersion string = ''
+param gpt4vDeploymentSkuName string = ''
+param gpt4vDeploymentCapacity int = 0
+var gpt4v = {
+  modelName: !empty(gpt4vModelName) ? gpt4vModelName : 'gpt-4o'
+  deploymentName: !empty(gpt4vDeploymentName) ? gpt4vDeploymentName : 'gpt-4o'
+  deploymentVersion: !empty(gpt4vModelVersion) ? gpt4vModelVersion : '2024-05-13'
+  deploymentSkuName: !empty(gpt4vDeploymentSkuName) ? gpt4vDeploymentSkuName : 'Standard'
+  deploymentCapacity: gpt4vDeploymentCapacity != 0 ? gpt4vDeploymentCapacity : 10
+}
 
 param tenantId string = tenant().tenantId
 param authTenantId string = ''
@@ -322,12 +335,12 @@ var appEnvVariables = {
   AZURE_OPENAI_EMB_MODEL_NAME: embedding.modelName
   AZURE_OPENAI_EMB_DIMENSIONS: embedding.dimensions
   AZURE_OPENAI_CHATGPT_MODEL: chatGpt.modelName
-  AZURE_OPENAI_GPT4V_MODEL: gpt4vModelName
+  AZURE_OPENAI_GPT4V_MODEL: gpt4v.modelName
   // Specific to Azure OpenAI
   AZURE_OPENAI_SERVICE: isAzureOpenAiHost && deployAzureOpenAi ? openAi.outputs.name : ''
   AZURE_OPENAI_CHATGPT_DEPLOYMENT: chatGpt.deploymentName
   AZURE_OPENAI_EMB_DEPLOYMENT: embedding.deploymentName
-  AZURE_OPENAI_GPT4V_DEPLOYMENT: useGPT4V ? gpt4vDeploymentName : ''
+  AZURE_OPENAI_GPT4V_DEPLOYMENT: useGPT4V ? gpt4v.deploymentName : ''
   AZURE_OPENAI_API_VERSION: azureOpenAiApiVersion
   AZURE_OPENAI_API_KEY_OVERRIDE: azureOpenAiApiKey
   AZURE_OPENAI_CUSTOM_URL: azureOpenAiCustomUrl
@@ -453,7 +466,7 @@ var defaultOpenAiDeployments = [
       version: chatGpt.deploymentVersion
     }
     sku: {
-      name: 'Standard'
+      name: chatGpt.deploymentSkuName
       capacity: chatGpt.deploymentCapacity
     }
   }
@@ -465,7 +478,7 @@ var defaultOpenAiDeployments = [
       version: embedding.deploymentVersion
     }
     sku: {
-      name: 'Standard'
+      name: embedding.deploymentSkuName
       capacity: embedding.deploymentCapacity
     }
   }
@@ -476,15 +489,15 @@ var openAiDeployments = concat(
   useGPT4V
     ? [
         {
-          name: gpt4vDeploymentName
+          name: gpt4v.deploymentName
           model: {
             format: 'OpenAI'
-            name: gpt4vModelName
-            version: gpt4vModelVersion
+            name: gpt4v.modelName
+            version: gpt4v.deploymentVersion
           }
           sku: {
-            name: 'Standard'
-            capacity: gpt4vDeploymentCapacity
+            name: gpt4v.deploymentSkuName
+            capacity: gpt4v.deploymentCapacity
           }
         }
       ]
@@ -960,15 +973,15 @@ output AZURE_RESOURCE_GROUP string = resourceGroup.name
 output OPENAI_HOST string = openAiHost
 output AZURE_OPENAI_EMB_MODEL_NAME string = embedding.modelName
 output AZURE_OPENAI_CHATGPT_MODEL string = chatGpt.modelName
-output AZURE_OPENAI_GPT4V_MODEL string = gpt4vModelName
+output AZURE_OPENAI_GPT4V_MODEL string = gpt4v.modelName
 
 // Specific to Azure OpenAI
 output AZURE_OPENAI_SERVICE string = isAzureOpenAiHost && deployAzureOpenAi ? openAi.outputs.name : ''
 output AZURE_OPENAI_API_VERSION string = isAzureOpenAiHost ? azureOpenAiApiVersion : ''
 output AZURE_OPENAI_RESOURCE_GROUP string = isAzureOpenAiHost ? openAiResourceGroup.name : ''
 output AZURE_OPENAI_CHATGPT_DEPLOYMENT string = isAzureOpenAiHost ? chatGpt.deploymentName : ''
 output AZURE_OPENAI_EMB_DEPLOYMENT string = isAzureOpenAiHost ? embedding.deploymentName : ''
-output AZURE_OPENAI_GPT4V_DEPLOYMENT string = isAzureOpenAiHost ? gpt4vDeploymentName : ''
+output AZURE_OPENAI_GPT4V_DEPLOYMENT string = isAzureOpenAiHost ? gpt4v.deploymentName : ''
 
 output AZURE_SPEECH_SERVICE_ID string = useSpeechOutputAzure ? speech.outputs.resourceId : ''
 output AZURE_SPEECH_SERVICE_LOCATION string = useSpeechOutputAzure ? speech.outputs.location : ''

infra/main.parameters.json

@@ -98,6 +98,9 @@
     "chatGptDeploymentVersion":{
       "value": "${AZURE_OPENAI_CHATGPT_DEPLOYMENT_VERSION}"
     },
+    "chatGptDeploymentSkuName":{
+      "value": "${AZURE_OPENAI_CHATGPT_DEPLOYMENT_SKU}"
+    },
     "chatGptDeploymentCapacity":{
       "value": "${AZURE_OPENAI_CHATGPT_DEPLOYMENT_CAPACITY}"
     },
@@ -110,14 +113,29 @@
     "embeddingDeploymentVersion":{
       "value": "${AZURE_OPENAI_EMB_DEPLOYMENT_VERSION}"
     },
+    "embeddingDeploymentSkuName":{
+      "value": "${AZURE_OPENAI_EMB_DEPLOYMENT_SKU}"
+    },
     "embeddingDeploymentCapacity":{
       "value": "${AZURE_OPENAI_EMB_DEPLOYMENT_CAPACITY}"
     },
     "embeddingDimensions": {
       "value": "${AZURE_OPENAI_EMB_DIMENSIONS}"
     },
+    "gpt4vModelName":{
+      "value": "${AZURE_OPENAI_GPT4V_MODEL}"
+    },
+    "gpt4vDeploymentName": {
+      "value": "${AZURE_OPENAI_GPT4V_DEPLOYMENT}"
+    },
+    "gpt4vDeploymentVersion":{
+      "value": "${AZURE_OPENAI_GPT4V_DEPLOYMENT_VERSION}"
+    },
+    "gpt4vDeploymentSkuName":{
+      "value": "${AZURE_OPENAI_GPT4V_DEPLOYMENT_SKU}"
+    },
     "gpt4vDeploymentCapacity":{
-      "value": "${AZURE_OPENAI_GPT4V_DEPLOYMENT_CAPACITY=10}"
+      "value": "${AZURE_OPENAI_GPT4V_DEPLOYMENT_CAPACITY}"
     },
     "openAiHost": {
       "value": "${OPENAI_HOST=azure}"