This document describes how to customize the deployment of the Agents Chat with Azure AI Foundry. Once you follow the steps here, you can run azd up
as described in the Deploying steps.
Disabling a resource will stop that resource from being created and deployed to your Azure Project.
- To disable AI Search, run
azd env set USE_SEARCH_SERVICE false
- To disable Application Insights, run
azd env set USE_APPLICATION_INSIGHTS false
- To disable Container Registry, run
azd env set USE_CONTAINER_REGISTRY false
Once you disable these resources, they will not be deployed when you run azd up
.
By default, this template will use a naming convention with unique strings to prevent naming collisions within Azure. To override default naming conventions, the following keys can be set:
AZURE_AIHUB_NAME
- The name of the AI Foundry Hub resourceAZURE_AIPROJECT_NAME
- The name of the AI Foundry ProjectAZURE_AIENDPOINT_NAME
- The name of the AI Foundry online endpoint used for deploymentsAZURE_AISERVICES_NAME
- The name of the Azure AI serviceAZURE_SEARCH_SERVICE_NAME
- The name of the Azure Search serviceAZURE_STORAGE_ACCOUNT_NAME
- The name of the Storage AccountAZURE_KEYVAULT_NAME
- The name of the Key VaultAZURE_CONTAINER_REGISTRY_NAME
- The name of the container registryAZURE_APPLICATION_INSIGHTS_NAME
- The name of the Application Insights instanceAZURE_LOG_ANALYTICS_WORKSPACE_NAME
- The name of the Log Analytics workspace used by Application Insights
To override any of those resource names, run azd env set <key> <value>
before running azd up
.
For more information on the Azure OpenAI models and non-Microsoft models that can be used in your deployment, view the list of models supported by Azure AI Agent Service.
To customize the model deployments, you can set the following environment variables:
Change the agent model format (either OpenAI or Microsoft):
azd env set AZURE_AI_AGENT_MODEL_FORMAT Microsoft
Change the agent model name:
azd env set AZURE_AI_AGENT_MODEL_NAME gpt-4o-mini
Set the version of the agent model:
azd env set AZURE_AI_AGENT_MODEL_VERSION 2024-07-18
For quota regions, you may find yourself needing to modify the default capacity and deployment SKU. The default tokens per minute deployed in this template is 50,000.
Change the capacity (in thousands of tokens per minute) of the agent deployment:
azd env set AZURE_AI_AGENT_DEPLOYMENT_CAPACITY 50
Change the SKU of the agent deployment:
azd env set AZURE_AI_AGENT_DEPLOYMENT_SKU Standard
Change the capacity (in thousands of tokens per minute) of the embeddings deployment:
azd env set AZURE_AI_EMBED_DEPLOYMENT_CAPACITY 50
Change the SKU of the embeddings deployment:
azd env set AZURE_AI_EMBED_DEPLOYMENT_SKU Standard