Azure OpenAI
Last updated
Was this helpful?
Azure OpenAI is Microsoft's OpenAI model service hosted on Azure, suitable for teams that need enterprise compliance, data residency commitments, or the Microsoft ecosystem.
Azure subscription activated
Applied for and approved for Azure OpenAI Service access in Azure Portal
At least one resource (Resource) and deployment (Deployment)
Azure Portal → Your Azure OpenAI resource → Keys and Endpoint
Copy KEY 1 or KEY 2, note down Endpoint(in the form of https://<your-name>.openai.azure.com/)
Open Settings → Model Services, find Azure OpenAI Provider to enter the details page
API key: paste in the copied KEY
API address: enter the resource Endpoint (no trailing path needed)
API Version: enter your deployment's API version in the dedicated field (for example 2024-08-01-preview)
Click Get model list, or manually add the model name you have deployed on Azure (i.e., Deployment Name, not the original OpenAI model ID)
Deployment Name vs Model ID: Azure uses the name you gave the deployment (e.g. gpt-4o-prod), not gpt-4o this original ID. Entering it wrong will cause a 404.
gpt-4o / gpt-4o-mini: general conversation, Agent
gpt-4 turbo: long context
text-embedding-3-*: embedding model, can be used with global memory, knowledge bases
401 Unauthorized: check whether the Key is correct, whether there is an extra slash at the end of the Endpoint
404 Not Found: check whether the Deployment Name matches Azure and whether the API Version was filled in
429 Throttled: check Azure quota (Quota & Limits page)
If you encounter any questions, bugs, or have suggestions for feature improvements during configuration or use, please refer to Feedback and Suggestions the official channels provided there.
Last updated
Was this helpful?
Was this helpful?