An Azure service that provides access to OpenAI’s GPT-3 models with enterprise capabilities.
Hi Nikita Khandare,
In Azure OpenAI, the key and endpoint you have are associated with the Azure OpenAI resource itself and are not tied to a specific model. The model that will be invoked is determined entirely by the deployment name (also called deployment_id) you provide in the API request. If you call the endpoint without specifying a deployment name, Azure OpenAI will not automatically choose one of your deployed models; instead, the request may either fail in strict API scenarios or return a success status without performing inference on any model. To ensure you are calling the correct model, always include the exact deployment name you assigned when creating the model deployment. Without this, the system cannot determine which of your deployed models to use. For more details, refer to Azure OpenAI – Deployments and Making requests to a deployment.