Planning model deployments: regions, model availability, deployment types (Standard, Global, Data Zone, Provisioned) and quotas

In Azure OpenAI you do not call a model directly. You create a deployment: a named instance of a specific model and version in your resource, with its own capacity. Your code calls the deployment name, not the model name, which lets you swap versions behind the same name. Planning a deployment means choosing a region, a model that is available there, a deployment type and how much quota to give it.