Azure OpenAI retirements from November 15 to December 16, 2026
The dataset has 8 Azure OpenAI ids in this window. 2026-11-15: codex-mini-2025-05-16 (Deprecated). 2026-11-17: gpt-chat-latest-2026-05-05 (Preview), gpt-chat-latest-2026-06-24 (Preview). 2026-12-08: gpt-chat-latest-2026-05-28 (Preview). 2026-12-15: tts-001 (Preview), tts-hd-001 (GA), whisper-001 (GA). 2026-12-16: gpt-image-1.5-2025-12-16 (GA). The retirement schedule's replacement cell is empty on each of those rows, so the table says Not published. The lifecycle page says a Retired model is removed from service and all inference requests return 410 Gone. Standard deployments can auto-upgrade. These rows do not name a replacement, so this page does not name an upgrade target. A deployment set to NoAutoUpgrade stops working. Provisioned deployments are not auto-upgraded.
Last reviewed . Publisher: SunsetDiff.
Steps
Read the date table. The table is these 8 Azure OpenAI ids. The lifecycle cell is the schedule's Lifecycle column, copied from the source excerpt stored on the row. Not published in the replacement cell means that field is empty. Open the model page before you change a deployment.
List deployments on the resource. Call Deployments List, or Get-AzCognitiveServicesAccountDeployment, for each Azure OpenAI resource. Read properties.model.name and properties.model.version. A deployment name can differ from the model name. codex-mini version 2025-05-16 is the dataset id codex-mini-2025-05-16. tts version 001 is tts-001.
Read versionUpgradeOption. OnceCurrentVersionExpired upgrades the deployment when the retirement date is reached. null means the same thing. NoAutoUpgrade never upgrades, and that deployment stops working. Provisioned deployments are not auto-upgraded.
Shortlist a model yourself. The schedule does not name a replacement for any row in the table. Do not wait for an automatic target. Pick a model, deploy it, and send one real request before the shutdown date on that row. The migration page asks: "Is there a named suggested replacement, or do you need to shortlist candidates yourself?"
Models and dates
These 8 ids are the Azure OpenAI rows whose shutdown date is 2026-11-15, 2026-11-17, 2026-12-08, 2026-12-15, or 2026-12-16. The November 19, 2026 o-series rows are a separate guide: Azure OpenAI o-series retirement on November 19, 2026. That page also mentions codex-mini and the two November 17 gpt-chat-latest versions.
The model id links to its SunsetDiff model page. Lifecycle is the schedule's Lifecycle column. Shutdown date and replacement are the fields stored on the row. Not published means the replacement field is empty. These rows do not store a deprecation_date. 2026-11-15: codex-mini-2025-05-16 (Deprecated). 2026-11-17: gpt-chat-latest-2026-05-05 (Preview), gpt-chat-latest-2026-06-24 (Preview). 2026-12-08: gpt-chat-latest-2026-05-28 (Preview). 2026-12-15: tts-001 (Preview), tts-hd-001 (GA), whisper-001 (GA). 2026-12-16: gpt-image-1.5-2025-12-16 (GA).
The schedule says: "The Replacement column identifies Microsoft's recommended replacement and, where supported, the target for automatic upgrades at retirement." This guide does not add a replacement the schedule does not print.
8 Azure OpenAI model ids retiring from 2026-11-15 through 2026-12-16.
The replacement cell on the retirement schedule is empty for every row in the table. Empty means Microsoft did not name a recommended replacement or an automatic-upgrade target on that row.
Other gpt-chat-latest rows in the dataset are not the replacement for the versions in the table. The schedule does not name them. gpt-chat-latest: Not published. gpt-chat-latest-2026-08-06: 2027-01-12 (/shutdowns/azure-openai-2027-01-12/). gpt-chat-latest-2026-10-07: 2027-05-18.
Other Azure OpenAI image rows in the dataset are not the replacement for gpt-image-1.5 2025-12-16. The schedule does not name them. gpt-image-1-2025-04-15: 2026-10-23. gpt-image-1-mini-2025-10-06: 2027-04-07. gpt-image-2-2026-04-21: 2027-10-21. gpt-image-2.5-flare-2026-09-09: 2027-09-09. gpt-image-2.5-sunburst-2026-09-09: 2027-09-09.
The dataset has no other tts, tts-hd, or whisper version. The schedule does not name a speech or transcription replacement for tts 001, tts-hd 001, or whisper 001. It also does not name a replacement for codex-mini 2025-05-16.
The migration page says: "Is there a named suggested replacement, or do you need to shortlist candidates yourself?" For these rows, the schedule does not name one, so you shortlist candidates yourself. It also says: "For each source model, note its retirement date and suggested replacement from the Model retirement schedule."
The lifecycle table says Retired means: "Removed from service. All inference requests return 410 Gone." New deployments are not allowed, and existing deployments do not work. The same page says the API value lifecycleStatus Deprecated means the model is retired and inference returns 410 Gone. The docs stage Deprecated is lifecycleStatus Deprecating, and inference still works. codex-mini 2025-05-16 is Deprecated on the schedule now, so inference still works until 2026-11-15. The sentences and the status map are on Azure OpenAI retired models return 410 Gone.
gpt-chat-latest 2026-05-05, gpt-chat-latest 2026-06-24, gpt-chat-latest 2026-05-28, and tts 001 are Preview on the schedule. The lifecycle page says: "Preview models aren't recommended for production workloads." It also says all preview deployments are either upgraded or terminated. And: "If no replacement exists, customers get 30 days notice before the model retires and inference returns 410 Gone." The schedule does not name a replacement, so this page does not pick the upgrade path or the no-replacement path for those preview rows.
tts-hd 001, whisper 001, and gpt-image-1.5 2025-12-16 are GA on the schedule. The lifecycle page says: "For Global Standard, Data Zone Standard, and Standard deployment types, Microsoft manages automatic upgrades when a model version is retired:" The migration page says: "Standard, Global Standard, and Data Zone Standard (pay-as-you-go) deployments are auto-upgraded on a rolling, region-by-region schedule." The replacement column is the target where an automatic upgrade is supported. These GA rows have an empty replacement cell, so this page does not name the model an auto-upgrade would select.
The lifecycle page says: "Provisioned deployments are NOT auto-upgraded. Provisioned customers must manually migrate to the replacement model." The migration page says: "Provisioned (PTU) deployments are not auto-upgraded." And: "Batch deployments follow the side-by-side path: deploy the new model, resubmit jobs, retire the old deployment."
Working with models says: "Automatic model updates are only supported for Standard deployment types." OnceCurrentVersionExpired means: "Once the retirement date is reached the model deployment automatically upgrades to the current default version." NoAutoUpgrade means: "The model deployment never automatically upgrades. Once the retirement date is reached the model deployment stops working." And: "null is equivalent to OnceCurrentVersionExpired."
The same page says: "When you select a specific model version for a deployment, this version remains selected until you either choose to manually update yourself, or once you reach the retirement date for the model." Then: "When the retirement date is reached the model will automatically upgrade to the default version at the time of retirement." A deployment you pinned to one of these versions can still move on the retirement date unless versionUpgradeOption is NoAutoUpgrade. NoAutoUpgrade does not move, and the deployment stops. That move still needs a default version. The schedule does not name one for these rows.
The migration page says: "You also need an inventory of the model deployments your workload uses, including their deployment types (Standard, Provisioned, or Batch)." Working with models says: "To query the current model deployment settings including the deployment upgrade configuration for a given resource use Deployments List." If the value is null, you won't see a versionUpgradeOption property. The model versions page says: "Go to your deployment in the Foundry portal, then open the Details tab to see the current model version." And: "Query the deployments endpoint for your resource. The response includes the model version for each deployment."
The example response puts the model name and version on properties.model. The deployment's own name is the name field next to that object. Match properties.model.name and properties.model.version to the ids in the table. gpt-chat-latest version 2026-05-28 is the dataset id gpt-chat-latest-2026-05-28. whisper version 001 is whisper-001. gpt-image-1.5 version 2025-12-16 is gpt-image-1.5-2025-12-16.
Also search the repository. A hit is a candidate, not proof a deployment still uses that version. How to find every model ID in your codebase covers SDK argument names. The scan page builds a search from dataset ids that have a shutdown date.
Azure PowerShell, list deployments for one account
GET https://management.azure.com/subscriptions/{subscriptionId}/resourceGroups/{resourceGroupName}/providers/Microsoft.CognitiveServices/accounts/{accountName}/deployments?api-version=2023-05-01
Create a deployment of a model you have tested, point traffic at it, and confirm one real request before the shutdown date on that row. Do that for each date: 2026-11-15, 2026-11-17, 2026-12-08, 2026-12-15, and 2026-12-16.
Working with models describes two ways to change a provisioned deployment once you have chosen a target. In-place migration keeps the deployment name and moves traffic. Multi-deployment migration creates a second deployment, shifts traffic, then deletes the old one. Neither section names a target for these rows.
The pre-migration test checklist is the longer list: evals, rejected fields, token limits, tools, price, and latency. These Azure rows do not store a price. The Azure OpenAI provider page lists the other stored rows.
Which Azure OpenAI versions retire between November 15 and December 16, 2026?
The dataset lists codex-mini-2025-05-16, gpt-chat-latest-2026-05-05, gpt-chat-latest-2026-06-24, gpt-chat-latest-2026-05-28, tts-001, tts-hd-001, whisper-001, gpt-image-1.5-2025-12-16. 2026-11-15: codex-mini-2025-05-16 (Deprecated). 2026-11-17: gpt-chat-latest-2026-05-05 (Preview), gpt-chat-latest-2026-06-24 (Preview). 2026-12-08: gpt-chat-latest-2026-05-28 (Preview). 2026-12-15: tts-001 (Preview), tts-hd-001 (GA), whisper-001 (GA). 2026-12-16: gpt-image-1.5-2025-12-16 (GA). The November 19 o-series ids are not in this list.
What is the official replacement?
The recommended replacement is the official replacement stored on each row. codex-mini-2025-05-16: Not published. gpt-chat-latest-2026-05-05: Not published. gpt-chat-latest-2026-06-24: Not published. gpt-chat-latest-2026-05-28: Not published. tts-001: Not published. tts-hd-001: Not published. whisper-001: Not published. gpt-image-1.5-2025-12-16: Not published. The schedule says: "The Replacement column identifies Microsoft's recommended replacement and, where supported, the target for automatic upgrades at retirement." The replacement cell is empty on each of these rows.
What does Azure return after retirement?
The lifecycle page says a Retired model is removed from service and: "Removed from service. All inference requests return 410 Gone." See Azure OpenAI retired models return 410 Gone.
Will every deployment upgrade itself?
No. The lifecycle page says: "For Global Standard, Data Zone Standard, and Standard deployment types, Microsoft manages automatic upgrades when a model version is retired:" The migration page says: "Standard, Global Standard, and Data Zone Standard (pay-as-you-go) deployments are auto-upgraded on a rolling, region-by-region schedule." And: "Provisioned (PTU) deployments are not auto-upgraded." Working with models says: "Automatic model updates are only supported for Standard deployment types." NoAutoUpgrade means: "The model deployment never automatically upgrades. Once the retirement date is reached the model deployment stops working." null is equivalent to OnceCurrentVersionExpired. These rows have no replacement on the schedule, so this page does not name an upgrade target.
How do I find deployments that still use these versions?
Working with models says: "To query the current model deployment settings including the deployment upgrade configuration for a given resource use Deployments List." Read properties.model.name and properties.model.version. The model versions page says: "Go to your deployment in the Foundry portal, then open the Details tab to see the current model version." The migration page says: "You also need an inventory of the model deployments your workload uses, including their deployment types (Standard, Provisioned, or Batch)."
Why is gpt-realtime-mini 2025-12-15 not in this table?
Checked on 2026-11-02, the retirement schedule prints gpt-realtime-mini version 2025-12-15 twice in the Azure OpenAI table. The first row says GA and retirement 2027-06-15. A later row says GA and retirement 2026-12-15. SunsetDiff keeps the first base-table row when the same model and version is repeated. The stored id is gpt-realtime-mini-2025-12-15 and its shutdown date is 2027-06-15. This guide skips that id. The June 15, 2027 rows are on Azure OpenAI gpt-4o-mini-tts, transcribe, and realtime audio retirement on June 15, 2027.
Does this guide include the November 19 o-series models?