Azure OpenAI

Azure OpenAI provides enterprise-grade access to OpenAI models through Microsoft Azure.

Setup

1. Set up Azure OpenAI in your Azure Portal
2. Deploy a model in Azure OpenAI Studio
3. Get your endpoint and API key
4. In Alma, go to Settings → Providers → Add Provider
5. Select Azure as the provider type
6. Configure your settings (see below)
7. Click Save

Configuration

Required Settings

Setting	Description	Example
Base URL	Your Azure endpoint	https://your-resource.openai.azure.com
API Key	Azure OpenAI key	abc123...
API Version	Azure API version	2024-08-01-preview

API Version

Use a recent API version for best compatibility. Common versions:

• 2024-08-01-preview - Recommended
• 2024-02-15-preview - Stable
• 2023-12-01-preview - Legacy

Responses API

Azure now supports a Responses API (similar to OpenAI's). Toggle this option if your deployment uses it.

Model Names

In Azure, model names are your deployment names, not the base model names. For example:

• If you deployed GPT-4 as "gpt4-deployment", use "gpt4-deployment"

Features

Enterprise Security

• VNet integration
• Private endpoints
• Azure AD authentication
• Compliance certifications

Regional Availability

Deploy models in Azure regions close to your users for lower latency.

Content Filtering

Azure provides configurable content filters for enterprise compliance.

Troubleshooting

"Resource Not Found"

• Check your endpoint URL includes the resource name
• Verify the deployment exists in Azure OpenAI Studio

"Authentication Failed"

• Ensure you're using the correct API key
• Check key expiration in Azure Portal

"Model Not Available"

• Verify the model is deployed in your Azure resource
• Check the deployment name matches exactly