Learn about Azure OpenAI Model Version Upgrades


Azure OpenAI Service is committed to providing the best generative models for customers. As part of this commitment, Azure OpenAI Service regularly releases new models to incorporate the latest features and improvements from OpenAI.  In 2023 we have introduced the following models widely used by customers:

  1. text-embedding-ada-002
  2. GPT-3.5-Turbo (versions 0301 and 0613)
  3. GPT-4 (versions 0314 and 0613)
  4. GPT-3.5-Turbo-Instruct
  5. babbage-002
  6. davinci-002

Of these models, GPT-3.5 Turbo and GPT-4 see regular updates with new features.  For example, versions 0613 of GPT-3.5 Turbo and GPT-4 introduced function calling, which is a popular feature that allows the model to create structured outputs that can be used to call external tools.  We will continue to see more capabilities with these models in the future.

We want to make it easy for customers to stay up to date as models improve.  Customers can choose to start with a particular version and to automatically update as new versions are released.

When a customer deploys GPT-3.5-Turbo and GPT-4 on Azure OpenAI Service, the standard behavior is to deploy the current default version – for example, GPT-4 version 0314.  When the default version changes to say GPT-4 version 0613, the deployment will be automatically updated to version 0613 so that customer deployments feature the latest capabilities of the model.

Customers can also deploy a specific version like GPT-4 0314 or GPT-4 0613 and choose an update policy, which can include auto-updating when the default version changes, auto-updating when the deployment's model version is retired, or never updating.  See below for links to the documentation.

Azure works closely with OpenAI to release new model versions.  When a new version of a model is released, a customer can immediately test it in new deployments.  Azure publishes when new versions of models are released and notifies customers at least two weeks before a new version becomes the default version of the model.   Azure also maintains the previous major version of the model until its retirement date, so customers can switch back to it if desired.

As a customer of Azure OpenAI models, you may notice some changes in the model behavior and compatibility after a version upgrade.  These changes may affect your applications and workflows that rely on the models.  For example, the response to prompts may change.  Here are some tips to help you prepare for version upgrades and minimize the impact:

  • Read the what's new and documentation of models to understand the changes and new features.
  • Read the documentation on model deployments and version upgrades to understand work with model versions.
  • Test your applications and workflows with the new model version after release.
  • Update your code and configuration to leverage the new features and capabilities of the new model version.
  • Looking for use case inspiration? Check out the most recent customer stories.
  • Need access to Azure OpenAI Service? Apply here.

Azure OpenAI Service models are constantly evolving and improving to provide the best and most reliable solutions for customers.  By following the Azure OpenAI Service model version upgrade policies, you can stay updated with the latest advancements and innovations from OpenAI and ensure a smooth transition for your applications and workflows.


This article was originally published by Microsoft's Azure AI Services Blog. You can find the original article here.