Skip to main content
Azure AI Foundry Models are continually refreshed with newer and more capable models. As part of this process, model providers might deprecate and retire their older models, and you might need to update your applications to use a newer model. This document communicates information about the model lifecycle and deprecation timelines and explains how you’re informed of model lifecycle stages. This article covers general deprecation and retirement information for Foundry Models. For details specific to Azure OpenAI in Foundry Models, see Azure OpenAI in Azure AI Foundry Models model deprecations and retirements.

​
Model lifecycle stages

Models in the model catalog belong to one of these stages:
  • Preview
  • Generally available
  • Legacy
  • Deprecated
  • Retired

​
Preview

Models labeled Preview are experimental in nature. A model’s weights, runtime, and API schema can change while the model is in preview. Models in preview aren’t guaranteed to become generally available. Models in preview have a Preview label next to their name in the model catalog.

​
Generally available

This stage is the default model stage. Models that don’t include a lifecycle label next to their name are generally available and suitable for use in production environments. In this stage, model weights and APIs are fixed. However, model containers or runtimes with vulnerabilities might get patched, but patches don’t affect model outputs.

​
Legacy

Models labeled Legacy are intended for deprecation. You should plan to move to a different model, such as a new, improved model that might be available in the same model family. While a model is in the legacy stage, existing deployments of the model continue to work, and you can create new deployments of the model until the deprecation date.

​
Deprecated

Models labeled Deprecated are no longer available for new deployments. You can’t create any new deployments for the model; however, existing deployments continue to work until the retirement date.

​
Retired

Models labeled Retired are no longer available for use. You can’t create new deployments, and attempts to use existing deployments return <return code> errors.

​
Notifications for Foundry Models

Customers that have Foundry Model deployments receive notifications for upcoming model retirements according to the following schedule:
  • Models are labeled as Legacy and remain in the legacy state for at least 30 days before being moved to the deprecated state. During this notification period, you can create new deployments as you prepare for deprecation and retirement.
  • Models are labeled Deprecated and remain in the deprecated state for at least 90 days before being moved to the retired state. During this notification period, you can migrate any existing deployments to newer or replacement models.
For each subscription that has a model deployed as a serverless API deployment or deployed to an Azure AI Foundry resource, members of the owner, contributor, reader, monitoring contributor, and monitoring reader roles receive a notification when a model deprecation is announced. The notification contains the dates when the model enters legacy, deprecated, and retired states. The notification might provide information about possible replacement model options, if applicable.

​
Notifications for Azure OpenAI in Foundry Models

For Azure OpenAI models, customers with active Azure OpenAI deployments receive notice for models with upcoming retirement as follows:
  • At model launch, we programmatically designate a “not sooner than” retirement date (typically one year out).
  • At least 60 days notice before model retirement for Generally Available (GA) models.
  • At least 30 days notice before preview model version upgrades.
Members of the owner, contributor, reader, monitoring contributor, and monitoring reader roles receive notification for each subscription with a deployment of a model that has an upcoming retirement. Retirements are done on a rolling basis, region by region. Notifications are sent from an unmonitored mailbox, azure-noreply@microsoft.com. To learn more about the Azure OpenAI models lifecycle, including information for current, deprecated, and retired models, see Azure OpenAI in Azure AI Foundry Models model deprecations and retirements.

​
Timelines for Foundry Models

The following tables list the timelines for models that are on track for retirement. The specified dates are in UTC time.

​
AI21 Labs

ModelLegacy date (UTC)Deprecation date (UTC)Retirement date (UTC)Suggested replacement model
Jamba InstructFebruary 1, 2025February 1, 2025March 1, 2025N/A
AI21-Jamba-1.5-LargeMay 1, 2025July 1, 2025August 1, 2025N/A
AI21-Jamba-1.5-MiniMay 1, 2025July 1, 2025August 1, 2025N/A

​
Cohere

ModelLegacy date (UTC)Deprecation date (UTC)Retirement date (UTC)Suggested replacement model
Command RFebruary 24, 2025March 25, 2025June 30, 2025Cohere Command R 08-2024
Command R+February 24, 2025March 25, 2025June 30, 2025Cohere Command R+ 08-2024
Cohere-rerank-v3-englishFebruary 28, 2025March 31, 2025June 30, 2025Cohere-rerank-v3.5
Cohere-rerank-v3-multilingualFebruary 28, 2025March 31, 2025June 30, 2025Cohere-rerank-v3.5

​
DeepSeek

ModelLegacy date (UTC)Deprecation date (UTC)Retirement date (UTC)Suggested replacement model
DeepSeek-V3April 10, 2025May 31, 2025August 31, 2025DeepSeek-V3-0324

​
Gretel

ModelLegacy date (UTC)Deprecation date (UTC)Retirement date (UTC)Suggested replacement model
Gretel-Navigator-TabularN/AJune 16, 2025September 16, 2025N/A

​
Meta

ModelLegacy date (UTC)Deprecation date (UTC)Retirement date (UTC)Suggested replacement model
Llama-2-13bFebruary 28, 2025March 31, 2025June 30, 2025Meta-Llama-3.1-8B-Instruct
Llama-2-13b-chatFebruary 28, 2025March 31, 2025June 30, 2025Meta-Llama-3.1-8B-Instruct
Llama-2-70bFebruary 28, 2025March 31, 2025June 30, 2025Llama-3.3-70B-Instruct
Llama-2-70b-chatFebruary 28, 2025March 31, 2025June 30, 2025Llama-3.3-70B-Instruct
Llama-2-7bFebruary 28, 2025March 31, 2025June 30, 2025Meta-Llama-3.1-8B-Instruct
Llama-2-7b-chatFebruary 28, 2025March 31, 2025June 30, 2025Meta-Llama-3.1-8B-Instruct
Meta-Llama-3-70B-InstructFebruary 28, 2025March 31, 2025June 30, 2025Llama-3.3-70B-Instruct
Meta-Llama-3-8B-InstructFebruary 28, 2025March 31, 2025June 30, 2025Meta-Llama-3.1-8B-Instruct
Meta-Llama-3.1-70B-InstructFebruary 28, 2025March 31, 2025June 30, 2025Llama-3.3-70B-Instruct

​
Microsoft

ModelLegacy date (UTC)Deprecation date (UTC)Retirement date (UTC)Suggested replacement model
Phi-3-medium-4k-instructJune 9, 2025June 30, 2025August 30, 2025Phi-4
Phi-3-medium-128k-instructJune 9, 2025June 30, 2025August 30, 2025Phi-4
Phi-3-mini-4k-instructJune 9, 2025June 30, 2025August 30, 2025Phi-4-mini-instruct
Phi-3-mini-128k-instructJune 9, 2025June 30, 2025August 30, 2025Phi-4-mini-instruct
Phi-3-small-8k-instructJune 9, 2025June 30, 2025August 30, 2025Phi-4-mini-instruct
Phi-3-small-128k-instructJune 9, 2025June 30, 2025August 30, 2025Phi-4-mini-instruct
Phi-3.5-mini-instructJune 9, 2025June 30, 2025August 30, 2025Phi-4-mini-instruct
Phi-3.5-MoE-instructJune 9, 2025June 30, 2025August 30, 2025Phi-4-mini-instruct
Phi-3.5-vision-instructJune 9, 2025June 30, 2025August 30, 2025Phi-4-mini-instruct

​
Mistral AI

ModelLegacy date (UTC)Deprecation date (UTC)Retirement date (UTC)Suggested replacement model
Mistral-smallMarch 31, 2025April 30, 2025July 31, 2025Mistral-small-2503
Mistral-large-2407January 13, 2025February 13, 2025May 13, 2025Mistral-large-2411
Mistral-largeDecember 15, 2024January 15, 2025April 15, 2025Mistral-large-2411

​
Related content