You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Copy file name to clipboardExpand all lines: articles/ai-foundry/agents/concepts/model-region-support.md
+17-20Lines changed: 17 additions & 20 deletions
Display the source diff
Display the rich diff
Original file line number
Diff line number
Diff line change
@@ -7,8 +7,8 @@ author: aahill
7
7
ms.author: aahi
8
8
ms.service: azure-ai-agent-service
9
9
ms.topic: conceptual
10
-
ms.date: 06/03/2025
11
-
ms.custom: azure-ai-agents
10
+
ms.date: 07/10/2025
11
+
ms.custom: azure-ai-agents, references_regions
12
12
---
13
13
14
14
# Models supported by Azure AI Foundry Agent Service
@@ -30,24 +30,21 @@ Azure AI Foundry Agent Service supports the following Azure OpenAI models in the
30
30
> * The following table is for serverless API deployment availability. For information on Provisioned Throughput Unit (PTU) availability, see [provisioned throughput](../../openai/concepts/provisioned-throughput.md) in the Azure OpenAI documentation. `GlobalStandard` customers also have access to [global standard models](../../openai/concepts/models.md#global-standard-model-availability).
31
31
> *[Hub based projects](../../what-is-azure-ai-foundry.md#project-types) are limited to the following models: gpt-4o, gpt-4o-mini, gpt-4, gpt-35-turbo
Copy file name to clipboardExpand all lines: articles/ai-foundry/agents/how-to/tools/deep-research.md
+1-1Lines changed: 1 addition & 1 deletion
Display the source diff
Display the rich diff
Original file line number
Diff line number
Diff line change
@@ -63,7 +63,7 @@ At its core, the Deep Research tool orchestrates a multi-step research pipeline
63
63
64
64
The Deep Research tool uses a second model deployment to clarify the question contained in the user prompt, gather additional context if needed, and precisely scope the research task. This helps make the output of an agent using the Deep Research tool more relevant and actionable, and can help optimize the search for your business scenario.
65
65
66
-
This second model is selected by the customer during configuration of the Deep Research tool and can be any of the GPT-series models including GPT-4o and GPT-4.1.
66
+
This second model is selected by the customer during configuration of the Deep Research tool and can be any of the GPT-series models including GPT-4o.
Copy file name to clipboardExpand all lines: articles/ai-foundry/concepts/foundry-models-overview.md
+12-15Lines changed: 12 additions & 15 deletions
Display the source diff
Display the rich diff
Original file line number
Diff line number
Diff line change
@@ -37,7 +37,7 @@ Understanding the distinction between these categories helps you choose the righ
37
37
38
38
## Models Sold Directly by Azure
39
39
40
-
These are models that are hosted and sold by Microsoft under Microsoft Product Terms. These models have undergone rigorous evaluation and are deeply integrated into Azure’s AI ecosystem. The models come from a variety of top providers and they offer enhanced integration, optimized performance, and direct Microsoft support, including enterprise-grade Service Level Agreements (SLAs).
40
+
These are models that are hosted and sold by Microsoft under Microsoft Product Terms. These models have undergone rigorous evaluation and are deeply integrated into Azure's AI ecosystem. The models come from a variety of top providers and they offer enhanced integration, optimized performance, and direct Microsoft support, including enterprise-grade Service Level Agreements (SLAs).
41
41
42
42
Characteristics of models sold directly by Azure:
43
43
@@ -51,7 +51,9 @@ These Models also have the benefit of fungible Provisioned Throughput, meaning y
51
51
52
52
## Models from Partners and Community
53
53
54
-
These models constitute the vast majority of the Azure AI Foundry Models. These models are provided by trusted third-party organizations, partners, research labs, and community contributors. These models offer specialized and diverse AI capabilities, covering a wide array of scenarios, industries, and innovations.
54
+
These models constitute the vast majority of the Azure AI Foundry Models and are provided by trusted third-party organizations, partners, research labs, and community contributors. These models offer specialized and diverse AI capabilities, covering a wide array of scenarios, industries, and innovations.
55
+
56
+
Examples of models from Partners and community are **Open models from the Hugging Face hub**. These include hundreds of models from the Hugging Face hub for real-time inference with managed compute. Hugging Face creates and maintains models listed in this collection. For help with the Hugging Face models, use the [Hugging Face forum](https://discuss.huggingface.co) or [Hugging Face support](https://huggingface.co/support). Learn how to deploy Hugging Face models in [Deploy open models with Azure AI Foundry](../how-to/deploy-models-managed.md).
55
57
56
58
Characteristics of Models from Partners and Community:
57
59
* Developed and supported by external partners and community contributors
@@ -60,26 +62,21 @@ Characteristics of Models from Partners and Community:
60
62
* Community-driven innovation and rapid availability of cutting-edge models
61
63
* Standard Azure AI integration, with support and maintenance managed by the respective providers
62
64
63
-
Models are deployable as Managed Compute or serverless API deployment options. The model provider selects how the models are deployable.
64
-
65
+
Models from Partners and Community are deployable as Managed Compute or serverless API deployment options. The model provider selects how the models are deployable.
66
+
67
+
### Requesting a model to be included in the model catalog
68
+
69
+
You can request that we add a model to the model catalog, right from the model catalog page in the Azure AI Foundry portal. From the search bar of the model catalog page, a search for a model that doesn't exist in the catalog, such as *mymodel*, returns the **Request a model** button. Select this button to open up a form where you can share details about the model you're requesting.
70
+
71
+
:::image type="content" source="../media/explore/model-request-button-in-catalog.png" alt-text="A screenshot showing where to request inclusion of a model in the model catalog." lightbox="../media/explore/model-request-button-in-catalog.png":::
72
+
65
73
## Choosing Between direct models and partner & community models
66
74
67
75
When selecting models from Azure AI Foundry Models, consider the following:
68
76
***Use Case and Requirements**: Models sold directly by Azure are ideal for scenarios requiring deep Azure integration, guaranteed support, and enterprise SLAs. Models from Partners and Community excel in specialized use cases and innovation-led scenarios.
69
77
***Support Expectations**: Models sold directly by Azure come with robust Microsoft-provided support and maintenance. These models are supported by their providers, with varying levels of SLA and support structures.
70
78
***Innovation and Specialization**: Models from Partners and Community offer rapid access to specialized innovations and niche capabilities often developed by leading research labs and emerging AI providers.
71
79
72
-
73
-
## Model collections
74
-
75
-
The model catalog organizes models into different collections, including:
76
-
77
-
***Azure OpenAI models exclusively available on Azure**: Flagship Azure OpenAI models available through an integration with Azure OpenAI in Foundry Models. Microsoft supports these models and their use according to the product terms and [SLA for Azure OpenAI](https://www.microsoft.com/licensing/docs/view/Service-Level-Agreements-SLA-for-Online-Services).
78
-
79
-
***Open models from the Hugging Face hub**: Hundreds of models from the Hugging Face hub for real-time inference with managed compute. Hugging Face creates and maintains models listed in this collection. For help, use the [Hugging Face forum](https://discuss.huggingface.co) or [Hugging Face support](https://huggingface.co/support). Learn more in [Deploy open models with Azure AI Foundry](../how-to/deploy-models-managed.md).
80
-
81
-
You can submit a request to add a model to the model catalog by using [this form](https://forms.office.com/pages/responsepage.aspx?id=v4j5cvGGr0GRqy180BHbR_frVPkg_MhOoQxyrjmm7ZJUM09WNktBMURLSktOWEdDODBDRjg2NExKUy4u).
82
-
83
80
## Overview of Model Catalog capabilities
84
81
85
82
The model catalog in Azure AI Foundry portal is the hub to discover and use a wide range of models for building generative AI applications. The model catalog features hundreds of models across model providers such as Azure OpenAI, Mistral, Meta, Cohere, NVIDIA, and Hugging Face, including models that Microsoft trained. Models from providers other than Microsoft are Non-Microsoft Products as defined in [Microsoft Product Terms](https://www.microsoft.com/licensing/terms/welcome/welcomepage) and are subject to the terms provided with the models.
![learn-build-service-prod[bot]](https://avatars.githubusercontent.com/in/237442?v=4&size=40){kind=avatar}
0 commit comments