Azure OpenAI gives you the same OpenAI models but deployed within your Azure environment, with enterprise compliance controls, private networking, and Azure-managed encryption.
Setup
You need an Azure OpenAI resource and a deployment.
export AZURE_OPENAI_API_KEY=...
export AZURE_OPENAI_ENDPOINT=https://my-resource.openai.azure.com
Config
{
"providers": {
"azure-openai-responses": {
"apiKey": "${AZURE_OPENAI_API_KEY}",
"baseUrl": "${AZURE_OPENAI_ENDPOINT}"
}
}
}
Use it
{
"agents": [
{ "name": "coder", "model": "azure-openai-responses:gpt-4o" }
]
}
The model ID must match the deployment name in your Azure OpenAI resource, not the base model name. If you deployed GPT-4o as my-gpt4o-deployment, use that as the model ID.
Models
Same models as OpenAI — availability depends on which models you’ve deployed in your Azure resource.
Features
| Feature | Supported |
|---|
| Streaming | Yes |
| Tool use | Yes |
| Vision (images) | Yes (GPT-4o deployments) |
| Private networking | Yes |
| Azure AD auth | Yes |
| Content filtering | Yes |
Provider Details
| |
|---|
| Provider ID | azure-openai-responses |
| Env variables | AZURE_OPENAI_API_KEY + AZURE_OPENAI_ENDPOINT |
| API type | Azure OpenAI Responses API |
| Base URL | Required (your Azure resource endpoint) |
When to Use Azure OpenAI vs. OpenAI Direct
| OpenAI Direct | Azure OpenAI |
|---|
| Auth | API key | API key or Azure AD |
| Compliance | SOC 2 | SOC 2, HIPAA, FedRAMP |
| Data residency | No control | Region-specific |
| Networking | Public | Private endpoints, VNet |
| Content filtering | Basic | Configurable |
| Best for | General use | Enterprise / regulated industries |
Notes
- Each model must be deployed separately in the Azure portal before it can be used.
- Azure OpenAI deployments have per-minute rate limits configured at the resource level.
- The
baseUrl must include the full endpoint URL (e.g. https://my-resource.openai.azure.com), not just the resource name.
