Skip to content

AZURE-OPENAI.md

Latest commit

 

History

History
88 lines (59 loc) · 2.05 KB

AZURE-OPENAI.md

File metadata and controls

88 lines (59 loc) · 2.05 KB
Raw

Azure OpenAI setup

How to configure the app to use Azure OpenAI directly.

When to use Azure OpenAI

Pick Azure OpenAI over Foundry when you already have AOAI resources, need a specific model version, or need regional deployment control for compliance or latency.

If you're starting fresh and don't have a strong reason to use AOAI directly, Foundry is usually simpler.

Compare providers

Prerequisites

Step 1: Create Azure OpenAI resource

# Navigate to the resource directory
cd resources-foundry

# Login to Azure
azd auth login

# Provision resources
azd up

Step 2: Get endpoint and API key

# Navigate to the resource directory
cd resources-foundry

# Login to Azure
az login

# Get endpoint
azd env get-value 'FOUNDRY_OPENAI_ENDPOINT'

# Get API key
az cognitiveservices account keys list -g rg-$(azd env get-value AZURE_ENV_NAME) -n $(azd env get-value FOUNDRY_NAME) --query "key1" -o tsv

Step 3: Store credentials

dotnet user-secrets --file ./apphost.cs set Azure:OpenAI:Endpoint "{{AZURE_OPENAI_ENDPOINT}}"
dotnet user-secrets --file ./apphost.cs set Azure:OpenAI:ApiKey "{{AZURE_OPENAI_API_KEY}}"

Step 4: Run the app

# Using file-based Aspire (recommended)
aspire run --file ./apphost.cs -- --provider AzureOpenAI

# Using project-based Aspire
aspire run --project ./src/InterviewCoach.AppHost -- --provider AzureOpenAI

Step 5: Deploy to Azure

# Login to Azure
azd auth login

# Provision + deploy
azd up

Step 6: Clean up

When finished, remove all Azure resources:

azd down --force --purge

Next steps