Skip to content

Commit f2e27ba

Browse files
mrbullwinklemrbullwinkle
committed
update
1 parent 5edcfa4 commit f2e27ba

File tree

3 files changed

+75
-317
lines changed

3 files changed

+75
-317
lines changed

‎articles/ai-services/openai/concepts/legacy-models.md

Lines changed: 50 additions & 121 deletions
Original file line numberDiff line numberDiff line change
@@ -1,133 +1,62 @@
11
---
2-
title: Azure OpenAI in Azure AI Foundry Models deprecated models
2+
title: Azure OpenAI in Azure AI Foundry Models retired models
33
titleSuffix: Azure OpenAI
44
description: Learn about the deprecated models in Azure OpenAI.
55
ms.service: azure-ai-openai
66
ms.topic: conceptual
7-
ms.date: 04/30/2025
7+
ms.date: 06/18/2025
88
ms.custom: references_regions, build-2023, build-2023-dataai
99
manager: nitinme
1010
author: mrbullwinkle
1111
ms.author: mbullwin
1212
recommendations: false
1313
---
1414

15-
# Azure OpenAI in Azure AI Foundry Models deprecated models
16-
17-
Azure OpenAI offers a variety of models for different use cases. The following models were deprecated on July 6, 2023 and will be retired on June 14, 2024. These models are no longer available for new deployments. Deployments created prior to July 6, 2023 remain available to customers until June 14, 2024. We recommend customers migrate their applications to deployments of replacement models prior to the June 14, 2024 retirement.
18-
19-
At the time of retirement, deployments of these models will stop returning valid API responses.
20-
21-
## GPT-3.5
22-
23-
The impacted GPT-3.5 models are the following. The replacement for the GPT-3.5 models is GPT-3.5 Turbo Instruct when that model becomes available.
24-
25-
- `text-davinci-002`
26-
- `text-davinci-003`
27-
- `code-davinci-002`
28-
29-
## GPT-3
30-
31-
The impacted GPT-3 models are the following. The replacement for the GPT-3 models is GPT-3.5 Turbo Instruct when that model becomes available.
32-
33-
- `text-ada-001`
34-
- `text-babbage-001`
35-
- `text-curie-001`
36-
- `text-davinci-001`
37-
- `code-cushman-001`
38-
39-
## Embedding models
40-
41-
The embedding models below will be retired effective June 14, 2024. Customers should migrate to `text-embedding-ada-002` (version 2).
42-
43-
- [Similarity](#similarity-embedding)
44-
- [Text search](#text-search-embedding)
45-
- [Code search](#code-search-embedding)
46-
47-
Each family includes models across a range of capability. The following list indicates the length of the numerical vector returned by the service, based on model capability:
48-
49-
| Base Model | Model(s) | Dimensions |
50-
|---|---|---|
51-
| Ada | | 1,024 |
52-
| Babbage | | 2,048 |
53-
| Curie | | 4,096 |
54-
| Davinci | | 12,288 |
55-
56-
57-
### Similarity embedding
58-
59-
These models are good at capturing semantic similarity between two or more pieces of text.
60-
61-
| Use cases | Models |
62-
|---|---|
63-
| Clustering, regression, anomaly detection, visualization | `text-similarity-ada-001` <br> `text-similarity-babbage-001` <br> `text-similarity-curie-001` <br> `text-similarity-davinci-001` <br>|
64-
65-
### Text search embedding
66-
67-
These models help measure whether long documents are relevant to a short search query. There are two input types supported by this family: `doc`, for embedding the documents to be retrieved, and `query`, for embedding the search query.
68-
69-
| Use cases | Models |
70-
|---|---|
71-
| Search, context relevance, information retrieval | `text-search-ada-doc-001` <br> `text-search-ada-query-001` <br> `text-search-babbage-doc-001` <br> `text-search-babbage-query-001` <br> `text-search-curie-doc-001` <br> `text-search-curie-query-001` <br> `text-search-davinci-doc-001` <br> `text-search-davinci-query-001` <br> |
72-
73-
### Code search embedding
74-
75-
Similar to text search embedding models, there are two input types supported by this family: `code`, for embedding code snippets to be retrieved, and `text`, for embedding natural language search queries.
76-
77-
| Use cases | Models |
78-
|---|---|
79-
| Code search and relevance | `code-search-ada-code-001` <br> `code-search-ada-text-001` <br> `code-search-babbage-code-001` <br> `code-search-babbage-text-001` |
80-
81-
## Model summary table and region availability
82-
83-
Region availability is for customers with deployments of the models prior to July 6, 2023.
84-
85-
### GPT-3.5 models
86-
87-
| Model ID | Base model Regions | Fine-Tuning Regions | Max Request (tokens) | Training Data (up to) |
88-
| --------- | --------------------- | ------------------- | -------------------- | ---------------------- |
89-
| text-davinci-002 | East US, South Central US, West Europe | N/A | 4,097 | Jun 2021 |
90-
| text-davinci-003 | East US, West Europe | N/A | 4,097 | Jun 2021 |
91-
| code-davinci-002 | East US, West Europe | N/A | 8,001 | Jun 2021 |
92-
93-
### GPT-3 models
94-
95-
96-
| Model ID | Base model Regions | Fine-Tuning Regions | Max Request (tokens) | Training Data (up to) |
97-
| --------- | --------------------- | ------------------- | -------------------- | ---------------------- |
98-
| ada | N/A | N/A | 2,049 | Oct 2019|
99-
| text-ada-001 | East US, South Central US, West Europe | N/A | 2,049 | Oct 2019|
100-
| babbage | N/A | N/A | 2,049 | Oct 2019 |
101-
| text-babbage-001 | East US, South Central US, West Europe | N/A | 2,049 | Oct 2019 |
102-
| curie | N/A | N/A | 2,049 | Oct 2019 |
103-
| text-curie-001 | East US, South Central US, West Europe | N/A | 2,049 | Oct 2019 |
104-
| davinci | N/A | N/A | 2,049 | Oct 2019|
105-
| text-davinci-001 | South Central US, West Europe | N/A | | |
106-
107-
108-
### Codex models
109-
110-
| Model ID | Base model Regions | Fine-Tuning Regions | Max Request (tokens) | Training Data (up to) |
111-
| --- | --- | --- | --- | --- |
112-
| code-cushman-001 | South Central US, West Europe | N/A | 2,048 | |
113-
114-
### Embedding models
115-
116-
| Model ID | Base model Regions | Fine-Tuning Regions | Max Request (tokens) | Training Data (up to) |
117-
| --- | --- | --- | --- | --- |
118-
| text-similarity-ada-001| East US, South Central US, West Europe | N/A | 2,046 | Aug 2020 |
119-
| text-similarity-babbage-001 | South Central US, West Europe | N/A | 2,046 | Aug 2020 |
120-
| text-similarity-curie-001 | East US, South Central US, West Europe | N/A | 2,046 | Aug 2020 |
121-
| text-similarity-davinci-001 | South Central US, West Europe | N/A | 2,046 | Aug 2020 |
122-
| text-search-ada-doc-001 | South Central US, West Europe | N/A | 2,046 | Aug 2020 |
123-
| text-search-ada-query-001 | South Central US, West Europe | N/A | 2,046 | Aug 2020 |
124-
| text-search-babbage-doc-001 | South Central US, West Europe | N/A | 2,046 | Aug 2020 |
125-
| text-search-babbage-query-001 | South Central US, West Europe | N/A | 2,046 | Aug 2020 |
126-
| text-search-curie-doc-001 | South Central US, West Europe | N/A | 2,046 | Aug 2020 |
127-
| text-search-curie-query-001 | South Central US, West Europe | N/A | 2,046 | Aug 2020 |
128-
| text-search-davinci-doc-001 | South Central US, West Europe | N/A | 2,046 | Aug 2020 |
129-
| text-search-davinci-query-001 | South Central US, West Europe | N/A | 2,046 | Aug 2020 |
130-
| code-search-ada-code-001 | South Central US, West Europe | N/A | 2,046 | Aug 2020 |
131-
| code-search-ada-text-001 | South Central US, West Europe | N/A | 2,046 | Aug 2020 |
132-
| code-search-babbage-code-001 | South Central US, West Europe | N/A | 2,046 | Aug 2020 |
133-
| code-search-babbage-text-001 | South Central US, West Europe | N/A | 2,046 | Aug 2020 |
15+
# Azure OpenAI in Azure AI Foundry Models retired models
16+
17+
Azure OpenAI offers a variety of models for different use cases. The following models are no longer available for deployment.
18+
19+
## Retired models
20+
21+
These models are no longer available for new deployments.
22+
23+
If you're an existing customer looking for information about these models, see [Legacy models](../../concepts/legacy-models.md).
24+
25+
| Model | Deprecation date | Retirement date | Suggested replacement |
26+
| --------- | --------------------- | ------------------- | -------------------- |
27+
| `gpt-4o-realtime-preview` - 2024-10-01 | February 25, 2025 | March 26, 2025 | `gpt-4o-realtime-preview` (version 2024-12-17) or `gpt-4o-mini-realtime-preview` (version 2024-12-17) |
28+
| `gpt-35-turbo` - 0301 | | February 13, 2025 | `gpt-35-turbo` (0125) <br><br> `gpt-4o-mini` |
29+
| `gpt-35-turbo` - 0613 | | February 13, 2025 | `gpt-35-turbo` (0125) <br><br> `gpt-4o-mini` |
30+
| `gpt-4`<br>`gpt-4-32k` - 0314 | | June 6, 2025 | `gpt-4o` version: `2024-11-20` |
31+
| `gpt-4`<br>`gpt-4-32k` - 0613 | | June 6, 2025 | `gpt-4o` version: `2024-11-20` |
32+
| `gpt-35-turbo-16k` - 0613 | | April 30, 2025 | `gpt-4.1-mini` version: `2025-04-14` |
33+
| `babbage-002` | | January 27, 2025 | |
34+
| `davinci-002` | | January 27, 2025 | |
35+
| `dall-e-2`| | January 27, 2025 | dalle-3 |
36+
| `ada` | July 6, 2023 | June 14, 2024 | |
37+
| `babbage` | July 6, 2023 | June 14, 2024 | |
38+
| `curie` | July 6, 2023 | June 14, 2024 | |
39+
| `davinci` | July 6, 2023 | June 14, 2024 | |
40+
| `text-ada-001` | July 6, 2023 | June 14, 2024 | gpt-35-turbo-instruct |
41+
| `text-babbage-001` | July 6, 2023 | June 14, 2024 | gpt-35-turbo-instruct |
42+
| `text-curie-001` | July 6, 2023 | June 14, 2024 | gpt-35-turbo-instruct |
43+
| `text-davinci-002` | July 6, 2023 | June 14, 2024 | gpt-35-turbo-instruct |
44+
| `text-davinci-003` | July 6, 2023 | June 14, 2024 | gpt-35-turbo-instruct |
45+
| `code-cushman-001` | July 6, 2023 | June 14, 2024 | gpt-35-turbo-instruct |
46+
| `code-davinci-002` | July 6, 2023 | June 14, 2024 | gpt-35-turbo-instruct |
47+
| `text-similarity-ada-001` | July 6, 2023 | June 14, 2024 | text-embedding-3-small |
48+
| `text-similarity-babbage-001` | July 6, 2023 | June 14, 2024 | text-embedding-3-small |
49+
| `text-similarity-curie-001` | July 6, 2023 | June 14, 2024 | text-embedding-3-small |
50+
| `text-similarity-davinci-001` | July 6, 2023 | June 14, 2024 | text-embedding-3-small |
51+
| `text-search-ada-doc-001` | July 6, 2023 | June 14, 2024 | text-embedding-3-small |
52+
| `text-search-ada-query-001` | July 6, 2023 | June 14, 2024 | text-embedding-3-small |
53+
| `text-search-babbage-doc-001` | July 6, 2023 | June 14, 2024 | text-embedding-3-small |
54+
| `text-search-babbage-query-001` | July 6, 2023 | June 14, 2024 | text-embedding-3-small |
55+
| `text-search-curie-doc-001` | July 6, 2023 | June 14, 2024 | text-embedding-3-small |
56+
| `text-search-curie-query-001` | July 6, 2023 | June 14, 2024 | text-embedding-3-small |
57+
| `text-search-davinci-doc-001` | July 6, 2023 | June 14, 2024 | text-embedding-3-small |
58+
| `text-search-davinci-query-001` | July 6, 2023 | June 14, 2024 | text-embedding-3-small |
59+
| `code-search-ada-code-001` | July 6, 2023 | June 14, 2024 | text-embedding-3-small |
60+
| `code-search-ada-text-001` | July 6, 2023 | June 14, 2024 | text-embedding-3-small |
61+
| `code-search-babbage-code-001` | July 6, 2023 | June 14, 2024 | text-embedding-3-small |
62+
| `code-search-babbage-text-001` | July 6, 2023 | June 14, 2024 | text-embedding-3-small |

‎articles/ai-services/openai/concepts/model-retirements.md

Lines changed: 19 additions & 141 deletions
Original file line numberDiff line numberDiff line change
@@ -26,15 +26,26 @@ Azure OpenAI models are continually refreshed with newer and more capable models
2626
* When a model is retired, it's no longer available for use. Azure OpenAI deployments of a retired model always return error responses.
2727

2828
## Notifications
29-
29+
3030
Azure OpenAI notifies customers of active Azure OpenAI deployments for models with upcoming retirements. We notify customers of upcoming retirements as follows for each deployment:
31-
31+
3232
1. At model launch, we programmatically designate a "not sooner than" retirement date (for preview models this is between 90-120 days from launch, for generally available (GA) models this is 365 days from launch).
3333
2. At least 60 days notice before model retirement for Generally Available (GA) models.
3434
3. At least 30 days notice before preview model version upgrades.
35-
35+
3636
Retirements are done on a rolling basis, region by region. There is no schedule for when a specific region, or SKU will be upgraded.
3737

38+
## Current models
39+
40+
> [!NOTE]
41+
> Not all models go through a deprecation period prior to retirement. Some models/versions only have a retirement date.
42+
>
43+
> **Fine-tuned models** are subject to a [different](#fine-tuned-models) deprecation and retirement schedule from their equivalent base model.
44+
45+
These models are currently available for use in Azure OpenAI.
46+
47+
[!INCLUDE [Model retirement table](../includes/retirement/models.md)]
48+
3849
## Model availability
3950

4051
1. At least one year of model availability for GA models after the release date of a model in at least one region worldwide
@@ -63,11 +74,13 @@ Be aware of the following:
6374
2. for model family changes beyond version updates, such as when moving from `gpt-4 1106-preview` to `gpt-4o (2024-05-13)`.
6475

6576
### Who is notified of upcoming retirements
66-
77+
6778
Azure OpenAI notifies customers via two methods:
68-
- **Azure Resoruce Health** - Anyone with reader permissions or above can see azure health alerts, as well as configure personalized alerts via email, SMS, etc. See [Create Service Health Alerts] (https://learn.microsoft.com/en-us/azure/service-health/alerts-activity-log-service-notifications-portal)
79+
- **Azure Resource Health** - Anyone with reader permissions or above can see Azure health alerts, as well as configure personalized alerts via email, SMS, etc. See [Create Service Health Alerts](/azure/service-health/alerts-activity-log-service-notifications-portal)
6980
- **Email** - email notifications are automatically sent to subscription owners. Any individual with reader permissions may however configure their own alerts by following the guidance above.
7081

82+
83+
7184
## How to get ready for model retirements and version upgrades
7285

7386
To prepare for model retirements and version upgrades, we recommend that customers test their applications with the new models and versions and evaluate their behavior. We also recommend that customers update their applications to use the new models and versions before the retirement date.
@@ -78,141 +91,6 @@ For information on the model upgrade process, see [How to upgrade to a new model
7891

7992
For more information on how to manage model upgrades and migrations for provisioned deployments, see [Managing models on provisioned deployment types](../how-to/working-with-models.md#managing-models-on-provisioned-deployment-types)
8093

81-
## Current models
82-
83-
> [!NOTE]
84-
> Not all models go through a deprecation period prior to retirement. Some models/versions only have a retirement date.
85-
>
86-
> **Fine-tuned models** are subject to a [different](#fine-tuned-models) deprecation and retirement schedule from their equivalent base model.
87-
88-
These models are currently available for use in Azure OpenAI.
89-
90-
[!INCLUDE [Model retirement table](../includes/retirement/models.md)]
91-
92-
9394
## Retirement and deprecation history
9495

95-
## April 15, 2025
96-
97-
To track further individual updates to this article refer to the [Git History](https://github.com/MicrosoftDocs/azure-ai-docs/commits/main/articles/ai-services/openai/concepts/model-retirements.md)
98-
99-
## March 18, 2025
100-
101-
GPT-4 preview models retirement date updated to date: May 1, 2025.
102-
103-
## February 28, 2025
104-
105-
- Updated `gpt-4` versions `1106-preview`, `0125-preview`, `vision-preview` to be upgraded to **`gpt-4o` version: `2024-11-20`**, starting no sooner than March 10, 2025.
106-
107-
## February 25, 2025
108-
109-
- `dalle-3` updated to no earlier than June 30, 2025.
110-
- `gpt-4o-realtime-preview` (2024-10-01) No earlier than March 26, 2025.
111-
112-
## February 20, 2025
113-
114-
- `o1-preview` updated to no earlier than April 2, 2025.
115-
116-
## February 12, 2025
117-
118-
- Updates to `gpt-35-turbo-16k` (0613), (1106), `gpt-35-turbo-instruct`, `gpt-4` (turbo-2024-04-09)
119-
120-
## February 6, 2025
121-
122-
- Updates to `gpt-35-turbo`, `gpt-4` preview models, and `gpt-4o` models.
123-
124-
## January 9, 2025
125-
126-
- `o1` added.
127-
- `gpt-35-turbo instruct` updated to no earlier than April 1, 2025.
128-
- `gpt-35-turbo` (0125) updated to no earlier than May 31, 2025.
129-
130-
## December 11, 2024
131-
132-
Embeddings models updated to no earlier than October 3, 2025.
133-
134-
## December 2, 2024
135-
136-
`gpt-3.5-turbo-instruct` updated to no earlier than February 1, 2025.
137-
138-
## November 22, 2024
139-
140-
`gpt-35-turbo` 1106 retirement date updated to no earlier than March 31, 2025.
141-
142-
## November 11, 2024
143-
144-
Updates to:
145-
146-
- `babbage-002`, `davinci-002`.
147-
- `gpt-35-turbo` DEFAULT model version update date.
148-
- `gpt-35-turbo` 0301, 0613 retirement date.
149-
- `gpt-35-turbo` 0125 retirement date.
150-
- `gpt-4o` DEFAULT model update date.
151-
- `text-embeddings-3-small` & `text-embedding-3-large` retirement date.
152-
153-
## October 25, 2024
154-
155-
* `babbage-002` & `davinci-002` deprecation date: November 15, 2024 and retirement date: January 27, 2025.
156-
157-
## September 12, 2024
158-
159-
* `gpt-35-turbo` (0301), (0613), (1106) and `gpt-35-turbo-16k` (0613) auto-update to default upgrade date updated to November 13, 2024.
160-
161-
## September 9, 2024
162-
163-
* `gpt-35-turbo` (0301) and (0613) retirement changed to January 27, 2025.
164-
* `gpt-4` preview model upgrade date changed to starting no sooner than January 27, 2025.
165-
166-
## September 3, 2024
167-
168-
* Updated tables to include information on `gpt-35-turbo` default version upgrades. Deployments of versions `0301`, `0613`, and `1106` set to [**Auto-update to default**](/azure/ai-services/openai/how-to/working-with-models?tabs=powershell#auto-update-to-default) will be automatically upgraded to version: `0125`, starting on November 15, 2024.|
169-
170-
### August 22, 2024
171-
172-
* Updated `gpt-35-turbo` (0301) retirement date to no earlier than November 1, 2024.
173-
* Updated `gpt4` and `gpt-4-32k` (0314 and 0613) deprecation date to November 1, 2024.
174-
175-
### August 8, 2024
176-
177-
* Updated `gpt-35-turbo` & `gpt-35-turbo-16k` (0613) model's retirement date to November 1, 2024.
178-
179-
### July 30, 2024
180-
181-
* Updated `gpt-4` preview model upgrade date to November 15, 2024 or later for the following versions:
182-
* 1106-preview
183-
* 0125-preview
184-
* vision-preview (Vision enhancements feature will no longer be supported once this model is retired/upgraded.)
185-
186-
### July 18, 2024
187-
188-
* Updated `gpt-4` 0613 deprecation date to October 1, 2024 and the retirement date to June 6, 2025.
189-
190-
### June 19, 2024
191-
192-
* Updated `gpt-35-turbo` 0301 retirement date to no earlier than October 1, 2024.
193-
* Updated `gpt-35-turbo` & `gpt-35-turbo-16k`0613 retirement date to October 1, 2024.
194-
* Updated `gpt-4` & `gpt-4-32k` 0314 deprecation date to October 1, 2024, and retirement date to June 6, 2025.
195-
196-
### June 4, 2024
197-
198-
Retirement date for legacy models updated by one month.
199-
200-
### April 24, 2024
201-
202-
Earliest retirement date for `gpt-35-turbo` 0301 and 0613 has been updated to August 1, 2024.
203-
204-
### March 13, 2024
205-
206-
We published this document to provide information about the current models, deprecated models, and upcoming retirements.
207-
208-
### February 23, 2024
209-
210-
We announced the upcoming in-place upgrade of `gpt-4` version `1106-preview` to `0125-preview` to start no earlier than March 8, 2024.
211-
212-
### November 30, 2023
213-
214-
The default version of `gpt-4` and `gpt-3-32k` was updated from `0314` to `0613` starting on November 30, 2023. The upgrade of `0314` deployments set for autoupgrade to `0613` was completed on December 3, 2023.
215-
216-
### July 6, 2023
217-
218-
We announced the deprecation of models with upcoming retirement on July 5, 2024.
96+
To track individual updates to this article refer to the [Git History](https://github.com/MicrosoftDocs/azure-ai-docs/commits/main/articles/ai-services/openai/includes/retirement/models.md)

0 commit comments

Comments
 (0)