Skip to content

Commit fffd4c5

Browse files
Albertyang0Albertyang0
authored
Merge pull request #1770 from MicrosoftDocs/main
OOB Publish 12/1 -- 5:00AM PST
2 parents 1754b71 + 110acd6 commit fffd4c5

File tree

8 files changed

+162
-5
lines changed

8 files changed

+162
-5
lines changed

‎articles/ai-studio/concepts/model-lifecycle-and-retirement.md

Lines changed: 75 additions & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -0,0 +1,75 @@
1+
---
2+
title: Deprecation and retirement for models in Azure AI model catalog
3+
titleSuffix: Azure AI Foundry
4+
description: Learn about the lifecycle stages, deprecation, and retirement for models in the Azure AI model catalog.
5+
manager: scottpolly
6+
ms.service: azure-ai-studio
7+
ms.topic: concept-article
8+
ms.date: 11/22/2024
9+
ms.author: mopeakande
10+
author: msakande
11+
ms.reviewer: kritifaujdar
12+
reviewer: fkriti
13+
14+
#Customer intent: As a data scientist, I want to learn about the lifecycle of models that are available in the model catalog.
15+
---
16+
17+
# Model deprecation and retirement in Azure AI model catalog
18+
19+
Models in the model catalog 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.
20+
21+
> [!IMPORTANT]
22+
> This article describes deprecation and retirement only for models that can be deployed to __serverless APIs__, not managed compute. To learn more about the differences between deployment to serverless APIs and managed computes, see [Model catalog and collections in Azure AI Foundry portal](../how-to/model-catalog-overview.md).
23+
24+
> [!NOTE]
25+
> Azure OpenAI models in the model catalog are provided through Azure OpenAI Service. For information about Azure Open AI model deprecation and retirement, see the [Azure OpenAI service product documentation](/azure/ai-services/openai/concepts/model-retirements).
26+
27+
## Model lifecycle stages
28+
29+
Models in the model catalog belong to one of these stages:
30+
31+
- Preview
32+
- Generally available
33+
- Legacy
34+
- Deprecated
35+
- Retired
36+
37+
### Preview
38+
39+
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.
40+
41+
### Generally available
42+
43+
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 won't affect model outputs.
44+
45+
### Legacy
46+
47+
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.
48+
49+
### Deprecated
50+
51+
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.
52+
53+
### Retired
54+
55+
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.
56+
57+
58+
## Notifications
59+
60+
- 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 may create new deployments as you prepare for deprecation and retirement.
61+
62+
- 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.
63+
64+
- Members of the _owner_, _contributor_, _reader_, monitoring contributor_, and _monitoring reader_ roles for each Azure subscription with a serverless API model deployment 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.
65+
66+
67+
68+
| Model provider | Model | Legacy date | Deprecation date | Retirement date | Suggested replacement model |
69+
| ---- | ---- | ---- | --- | ---- | --- |
70+
| Mistral AI | [Mistral-large](https://aka.ms/azureai/landing/Mistral-Large) | December 15, 2024 | January 15, 2025 | April 15, 2025 | [Mistral-large-2407](https://aka.ms/azureai/landing/Mistral-Large-2407) |
71+
72+
## Related content
73+
74+
- [Model catalog and collections in Azure AI Foundry portal](../how-to/model-catalog-overview.md)
75+
- [Data, privacy, and security for use of models through the model catalog in AI Foundry portal](../how-to/concept-data-privacy.md)

‎articles/ai-studio/how-to/deploy-models-mistral.md

Lines changed: 2 additions & 2 deletions
Original file line numberDiff line numberDiff line change
@@ -2213,9 +2213,9 @@ For more information on how to track costs, see [Monitor costs for models offere
22132213
22142214
## Related content
22152215
2216-
2217-
* [Azure AI Model Inference API](../reference/reference-model-inference-api.md)
22182216
* [Deploy models as serverless APIs](deploy-models-serverless.md)
2217+
* [Azure AI Model Inference API](../reference/reference-model-inference-api.md)
2218+
* [Model deprecation and retirement in Azure AI model catalog](../concepts/model-lifecycle-and-retirement.md)
22192219
* [Consume serverless API endpoints from a different Azure AI Foundry project or hub](deploy-models-serverless-connect.md)
22202220
* [Region availability for models in serverless API endpoints](deploy-models-serverless-availability.md)
22212221
* [Plan and manage costs (marketplace)](costs-plan-manage.md#monitor-costs-for-models-offered-through-the-azure-marketplace)

‎articles/ai-studio/how-to/model-catalog-overview.md

Lines changed: 1 addition & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -188,3 +188,4 @@ To set the public network access flag for the AI Foundry hub:
188188
## Related content
189189

190190
* [Explore foundation models in Azure AI Foundry portal](models-foundation-azure-ai.md)
191+
* [Model deprecation and retirement in Azure AI model catalog](../concepts/model-lifecycle-and-retirement.md)

‎articles/ai-studio/toc.yml

Lines changed: 4 additions & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -91,6 +91,10 @@ items:
9191
href: how-to/model-catalog-overview.md
9292
- name: Data, privacy, and security for Model Catalog
9393
href: how-to/concept-data-privacy.md
94+
- name: Model lifecycle and retirement
95+
href: concepts/model-lifecycle-and-retirement.md
96+
- name: Model benchmarks
97+
href: how-to/model-benchmarks.md
9498
- name: Model benchmarking
9599
items:
96100
- name: Model benchmarks

‎articles/machine-learning/concept-model-catalog.md

Lines changed: 2 additions & 3 deletions
Original file line numberDiff line numberDiff line change
@@ -165,6 +165,5 @@ To set the PNA flag for the workspace:
165165

166166
## Learn more
167167

168-
* Learn [how to use foundation Models in Azure Machine Learning](./how-to-use-foundation-models.md) for fine-tuning, evaluation, and deployment using Azure Machine Learning studio UI or code based methods.
169-
* Explore the [Model Catalog in Azure Machine Learning studio](https://ml.azure.com/model/catalog). You need an [Azure Machine Learning workspace](./quickstart-create-resources.md) to explore the catalog.
170-
* [Evaluate, fine tune, and deploy models](./how-to-use-foundation-models.md) curated by Azure Machine Learning.
168+
- [Model deprecation and retirement in Azure AI model catalog](concept-model-lifecycle-and-retirement.md)
169+
- [How to use Open Source foundation models curated by Azure Machine Learning](how-to-use-foundation-models.md)

‎articles/machine-learning/concept-model-lifecycle-and-retirement.md

Lines changed: 75 additions & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -0,0 +1,75 @@
1+
---
2+
title: Deprecation and retirement for models in Azure AI model catalog
3+
titleSuffix: Azure Machine Learning
4+
description: Learn about the lifecycle stages, deprecation, and retirement for models in the Azure AI model catalog.
5+
services: machine-learning
6+
ms.service: azure-machine-learning
7+
ms.subservice: inferencing
8+
ms.topic: conceptual
9+
ms.author: mopeakande
10+
author: msakande
11+
ms.reviewer: kritifaujdar
12+
reviewer: fkriti
13+
ms.date: 11/22/2024
14+
#Customer intent: As a data scientist, I want to learn about the lifecycle of models that are available in the model catalog.
15+
---
16+
17+
# Deprecation and retirement of models in Azure AI model catalog
18+
19+
Models in the model catalog 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.
20+
21+
> [!IMPORTANT]
22+
> This article describes deprecation and retirement only for models that can be deployed to __serverless APIs__, not managed compute. To learn more about the differences between deployment to serverless APIs and managed computes, see [Model Catalog and Collections](concept-model-catalog.md).
23+
24+
> [!NOTE]
25+
> Azure OpenAI models in the model catalog are provided through Azure OpenAI Service. For information about Azure Open AI model deprecation and retirement, see the [Azure OpenAI service product documentation](/azure/ai-services/openai/concepts/model-retirements).
26+
27+
## Model lifecycle stages
28+
29+
Models in the model catalog belong to one of these stages:
30+
31+
- Preview
32+
- Generally available
33+
- Legacy
34+
- Deprecated
35+
- Retired
36+
37+
### Preview
38+
39+
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.
40+
41+
### Generally available
42+
43+
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 won't affect model outputs.
44+
45+
### Legacy
46+
47+
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.
48+
49+
### Deprecated
50+
51+
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.
52+
53+
### Retired
54+
55+
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.
56+
57+
58+
## Notifications
59+
60+
- 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 may create new deployments as you prepare for deprecation and retirement.
61+
62+
- 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.
63+
64+
- Members of the _owner_, _contributor_, _reader_, monitoring contributor_, and _monitoring reader_ roles for each Azure subscription with a serverless API model deployment 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.
65+
66+
67+
68+
| Model provider | Model | Legacy date | Deprecation date | Retirement date | Suggested replacement model |
69+
| ---- | ---- | ---- | --- | ---- | --- |
70+
| Mistral AI | [Mistral-large](https://aka.ms/azureai/landing/Mistral-Large) | December 15, 2024 | January 15, 2025 | April 15, 2025 | [Mistral-large-2407](https://aka.ms/azureai/landing/Mistral-Large-2407) |
71+
72+
## Related content
73+
74+
- [Model Catalog and Collections](concept-model-catalog.md)
75+
- [Data, privacy, and security for use of models through the Model Catalog](concept-data-privacy.md)

‎articles/machine-learning/how-to-deploy-models-mistral.md

Lines changed: 1 addition & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -329,6 +329,7 @@ Models deployed as a service with pay-as-you-go are protected by Azure AI conten
329329
## Related content
330330

331331
- [Model Catalog and Collections](concept-model-catalog.md)
332+
- [Model deprecation and retirement in Azure AI model catalog](concept-model-lifecycle-and-retirement.md)
332333
- [Region availability for models in serverless API endpoints](concept-endpoint-serverless-availability.md)
333334
- [Deploy and score a machine learning model by using an online endpoint](how-to-deploy-online-endpoints.md)
334335
- [Plan and manage costs for Azure AI Foundry](/azure/ai-studio/how-to/costs-plan-manage)

‎articles/machine-learning/toc.yml

Lines changed: 2 additions & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -590,6 +590,8 @@ items:
590590
href: concept-model-catalog.md
591591
- name: Data, privacy, and security for Model Catalog
592592
href: concept-data-privacy.md
593+
- name: Model lifecycle and retirement
594+
href: concept-model-lifecycle-and-retirement.md
593595
- name: Open source models curated by Azure Machine Learning
594596
href: how-to-use-foundation-models.md
595597
- name: Hugging Face Hub community partner models

0 commit comments

Comments
 (0)