How to deploy gpt4o-mini on Microsoft Azure and get a "Custom LLM URL" for integration?

Question

I’m trying to deploy gpt4o-mini on Microsoft Azure to get a URL I can use as a "Custom LLM URL" in my AI application.

I’ve followed the documentation, created an Azure OpenAI resource, and tested the model in the playground. However, I can’t figure out how to expose the model as an endpoint or retrieve the URL for external use.

How can I deploy gpt4o-mini and get the endpoint URL? Am I missing a key step?

Any help would be greatly appreciated!

Thanks!

Reviewed all relevant Microsoft Azure documentation. Tested the model in the playground, but the URL doesn’t seem to work. I feel like I might be missing something fundamental.

Answer 1

If you are looking for the azure openai service endpoint for your model deployment, you can find it Azure AI Foundry. Go to deployments and then select the model deployment. in the endpoint section, the Target URI and Key are what you need to use in your application to invoke LLM calls.

In case you need an example of making the call, in the playground, there is a View code option that will give you a code block that you can use. The same endpoint url and key are used in this sample code.