Azure OpenAI

Azure OpenAI is the OpenAI model service hosted by Microsoft on Azure, suitable for teams requiring enterprise compliance, data residency commitments, or integration with the Microsoft ecosystem.

Prerequisites

An active Azure subscription
Access approval for Azure OpenAI Service obtained via Azure Portal
At least one Resource and Deployment created

Obtain API Key

Azure Portal → Your Azure OpenAI resource → Keys and Endpoint
Copy KEY 1 or KEY 2, and note down Endpoint (formatted as https://<your-name>.openai.azure.com/)

Configure in BLACK HOLE

Open Settings → Model Provider, locate the Azure OpenAI Provider, and enter the details page
API Key: Enter the copied KEY
API Endpoint: Enter the resource Endpoint (without the trailing path)
API Version: Enter the API version of your deployment in the dedicated field (e.g., 2024-08-01-preview)
Click Fetch Model List, or manually add the model names you have deployed on Azure (i.e., the Deployment Name, not the original OpenAI model ID)

Recommended Usage

gpt-4o / gpt-4o-mini: General conversation, Agent
gpt-4 turbo: Long context
text-embedding-3-*: Embedding models, usable for knowledge bases

FAQ

401 Unauthorized: Verify the Key is correct and check for unnecessary trailing slashes in the Endpoint
404 Not Found: Verify the Deployment Name matches Azure exactly and that the API Version is filled in
429 Throttled: Check Azure quotas (Quota & Limits page)

Get Help and Submit Feedback

If you have any questions, encounter bugs, or have feature improvement suggestions during configuration or usage, please refer to the official channels provided in Feedback and Suggestions.