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/how-to/tools/bing-custom-search.md
+2-2Lines changed: 2 additions & 2 deletions
Display the source diff
Display the rich diff
Original file line number
Diff line number
Diff line change
@@ -7,7 +7,7 @@ manager: nitinme
7
7
ms.service: azure-ai-foundry
8
8
ms.subservice: azure-ai-foundry-agent-service
9
9
ms.topic: how-to
10
-
ms.date: 08/15/2025
10
+
ms.date: 09/26/2025
11
11
author: aahill
12
12
ms.author: aahi
13
13
ms.custom: azure-ai-agents
@@ -40,7 +40,7 @@ The authorization will happen between Grounding with Bing Custom Search service
40
40
41
41
Developers and end users don't have access to raw content returned from Grounding with Bing Custom Search. The model response, however, includes citations with links to the websites used to generate the response and is allowed to be stored using the mechanisms provided by the Agents Service. You can retrieve the model response by accessing the data in the thread that was created. These references must be retained and displayed in the exact form provided by Microsoft, as per Grounding with Bing Custom Search's Use and Display Requirements.
42
42
43
-
43
+
Transactions with your Grounding with Bing resource are counted by the number of tool calls per run. You can see how many tool calls are made from the run step.
Copy file name to clipboardExpand all lines: articles/ai-foundry/agents/how-to/tools/bing-grounding.md
+3-1Lines changed: 3 additions & 1 deletion
Display the source diff
Display the rich diff
Original file line number
Diff line number
Diff line change
@@ -7,7 +7,7 @@ manager: nitinme
7
7
ms.service: azure-ai-foundry
8
8
ms.subservice: azure-ai-foundry-agent-service
9
9
ms.topic: how-to
10
-
ms.date: 08/07/2025
10
+
ms.date: 09/26/2025
11
11
author: aahill
12
12
ms.author: aahi
13
13
ms.custom: azure-ai-agents
@@ -38,6 +38,8 @@ Grounding with Bing returns relevant search results to the customer's model depl
38
38
39
39
The authorization will happen between Grounding with Bing Search service and Azure AI Foundry Agent Service. Any Bing search query that is generated and sent to Bing for the purposes of grounding is transferred, along with the resource key, outside of the Azure compliance boundary to the Grounding with Bing Search service. Grounding with Bing Search is subject to Bing's terms and do not have the same compliance standards and certifications as the Azure AI Foundry Agent Service, as described in the [Grounding with Bing Search Terms of Use](https://www.microsoft.com/bing/apis/grounding-legal). It is your responsibility to assess whether the use of Grounding with Bing Search in your agent meets your needs and requirements.
40
40
41
+
Transactions with your Grounding with Bing resource are counted by the number of tool calls per run. You can see how many tool calls are made from the run step.
42
+
41
43
## Supported capabilities and known issues
42
44
- Grounding with Bing Search tool is designed to retrieve real-time information from web, NOT specific web domains.
43
45
- NOT Recommended to **summarize** an entire web page.
Copy file name to clipboardExpand all lines: articles/ai-foundry/agents/overview.md
+15-1Lines changed: 15 additions & 1 deletion
Display the source diff
Display the rich diff
Original file line number
Diff line number
Diff line change
@@ -8,7 +8,7 @@ ms.author: aahi
8
8
ms.service: azure-ai-foundry
9
9
ms.subservice: azure-ai-foundry-agent-service
10
10
ms.topic: overview
11
-
ms.date: 06/25/2025
11
+
ms.date: 09/26/2025
12
12
ms.custom: azure-ai-agents
13
13
---
14
14
@@ -117,6 +117,20 @@ Start with the [environment setup](environment-setup.md) and [quickstart](quicks
117
117
1. After you create a project, you can deploy a compatible model such as GPT-4o.
118
118
1. When you have a deployed model, you can also start making API calls to the service using the SDKs.
119
119
120
+
## Business Continuity and Disaster Recovery (BCDR) for Agents
121
+
122
+
To support service resilience, the Azure AI Foundry Agent service relies on customer-provisioned Cosmos DB accounts. This ensures that your agent state can be preserved and recovered in the event of a regional outage.
123
+
124
+
### Use your own Cosmos DB account
125
+
126
+
* As an Azure Standard customer, you provision and manage your own single-tenant Cosmos DB account. All agent state is stored in your Cosmos DB.
127
+
* Backup and recovery rely on Cosmos DB’s native capabilities, which you control.
128
+
* If the primary region becomes unavailable, the agent will automatically become available in the secondary region by connecting to the same Cosmos DB account.
129
+
* Since all history is preserved in Cosmos DB, the agent can continue operation with minimal disruption.
130
+
131
+
### Current guidance
132
+
133
+
We recommend customers provision and maintain their Cosmos DB account and ensure appropriate backup and recovery policies are configured. This ensures seamless continuity if the primary region becomes unavailable.
Copy file name to clipboardExpand all lines: articles/ai-foundry/concepts/resource-types.md
+13-4Lines changed: 13 additions & 4 deletions
Display the source diff
Display the rich diff
Original file line number
Diff line number
Diff line change
@@ -20,20 +20,29 @@ An Azure resource is required to use and manage services in Azure. It defines th
20
20
21
21
Azure AI Foundry portal and SDK clients support multiple distinct Azure resource types, each designed to serve different development and operational needs. This article explains which use case requires which type.
22
22
23
-
## Resource Types supported with AI Foundry
23
+
## Resource Types supported with AI Foundry Portal and SDK clients
24
24
25
-
***Azure AI Foundry** – The primary resource type for designing, deploying, and managing generative AI applications and agents. It provides access to agent service, models that are hosted using a serverless hosting model, evaluations, and Azure OpenAI service. Azure AI Foundry is the recommended resource type for most applications built in Azure AI Foundry.
25
+
***Azure AI Foundry** – The primary resource type for designing, deploying, and managing generative AI applications and agents. It provides access to the superset of Azure AI capabilities. This includes agent service, models sold by Microsoft and its partners, evaluations, AI Services and Azure OpenAI service capabilities. Azure AI Foundry is the recommended resource type for most applications built in Azure AI Foundry.
26
26
27
27
Get started by [creating a first AI Foundry resource](../../ai-services/multi-service-resource.md?context=/azure/ai-foundry/context/context).
28
28
29
-
***Azure AI Hub** – Use this resource type in combination with Azure AI Foundry to additionally access open-source model hosting and fine-tuning capabilities, as well as Azure Machine Learning capabilities. When you create an AI Hub, an Azure AI Foundry resource is automatically provisioned. Hub resources can be used in both Azure AI Foundry portal and Machine Learning Studio.
29
+
:::image type="content" source="../media/concepts/foundry-resource.png" alt-text="Screenshot shows how AI Foundry resource type provided access to the superset of Azure AI capabilities including agent service, a wide selection of models, and Azure OpenAI capabilities.":::
30
30
31
31
***Azure AI Search** – A resource used to index and retrieve data for grounding AI applications. It can be [connected](../how-to/connections-add.md) to Azure AI Foundry agents to enable retrieval-augmented generation (RAG) and semantic search experiences.
32
32
33
-
***Azure OpenAI** – A specialized resource type that provides access to OpenAI models such as GPT-4 and GPT-4o. It offers a subset of the capabilities available in Azure AI Foundry and provides solely access to Azure OpenAI APIs. [Upgrade from Azure OpenAI to Azure AI Foundry](../how-to/upgrade-azure-openai.md) to gain access to more capabilities while keeping the existing Azure OpenAI API endpoint, state of work, and security configurations.
33
+
***Azure OpenAI** – A specialized resource type that provides access to OpenAI models and APIs only. For most use cases, it's recommended to use the Azure AI Foundry resource, which offers backwards compatibility with all Azure OpenAI APIs.
34
+
35
+
If your IT security team hasn't enabled the superset of Foundry capabilities in your environment, then the standalone Azure OpenAI resource may still be required for you.
36
+
37
+
[An upgrade option from Azure OpenAI to AI Foundry](../how-to/upgrade-azure-openai.md) is available to access all Foundry capabilities and models while keeping your existing Azure OpenAI API endpoint, state of work, and security configurations.
38
+
39
+
***Azure AI Hub** - In June 2025, we started to move most of Hub's capabilities under "Azure AI Foundry" resource type. This change brings agents, models, and their tools together for development, management and governance, under a dedicated Azure resource type for AI Foundry.
40
+
41
+
New features will only release on Azure AI Foundry resource. To learn more, see [migrate from hub-based to Foundry projects](../how-to/migrate-project.md). [Select use cases](../what-is-azure-ai-foundry.md#which-type-of-project-do-i-need), including open source model deployments, currently still require a hub resource.
34
42
35
43
## References
36
44
45
+
*[Azure AI Foundry architecture](architecture.md)
37
46
*[What is Azure Resource Manager?](/azure/azure-resource-manager/management/overview)
38
47
*[Create a first AI Foundry resource](../../ai-services/multi-service-resource.md?context=/azure/ai-foundry/context/context)
39
48
*[Create AI Foundry with advanced options](../how-to/create-resource-template.md)
Copy file name to clipboardExpand all lines: articles/ai-foundry/foundry-models/includes/models-azure-direct-others.md
+5-6Lines changed: 5 additions & 6 deletions
Display the source diff
Display the rich diff
Original file line number
Diff line number
Diff line change
@@ -4,7 +4,7 @@ manager: nitinme
4
4
ms.service: azure-ai-foundry
5
5
ms.subservice: azure-ai-foundry-model-inference
6
6
ms.topic: include
7
-
ms.date: 09/17/2025
7
+
ms.date: 09/26/2025
8
8
ms.author: mopeakande
9
9
author: msakande
10
10
---
@@ -16,7 +16,7 @@ author: msakande
16
16
17
17
The Black Forest Labs (BFL) collection of image generation models includes FLUX.1 Kontext [pro] for in-context generation and editing and FLUX1.1 [pro] for text-to-image generation.
18
18
19
-
You can run these models via the BFL service provider API and through the [images/generations and images/edits endpoints](../../openai/reference-preview.md).
19
+
You can run these models through the BFL service provider API and through the [images/generations and images/edits endpoints](../../openai/reference-preview.md).
20
20
21
21
22
22
| Model | Type & API endpoint| Capabilities | Deployment type (region availability) | Project type |
@@ -82,12 +82,11 @@ xAI's Grok models in Azure AI Foundry Models include a diverse set of models des
82
82
83
83
- Grok 3 Mini is a lightweight reasoning model trained to tackle agentic, coding, mathematical, and deep science problems with test-time compute. It also supports a 131,072 token context window for understanding codebases and enterprise documents, and excels at using tools to solve complex logical problems in novel environments, offering raw reasoning traces for user inspection with adjustable thinking budgets.
84
84
85
-
- Grok Code Fast 1, a fast and efficient reasoning model designed for use in agentic coding applications. It was pre-trained on a coding-focused data mixture, then post-trained on demonstrations of various coding tasks and tool use as well as demonstrations of correct refusal behaviors based on xAI's safety policy.
86
-
87
-
> [!NOTE]
88
-
> [Registration is required for access to the grok-code-fast-1 model](https://aka.ms/xai/grok-code-fast-1).
85
+
- Grok Code Fast 1, a fast and efficient reasoning model designed for use in agentic coding applications. It was pretrained on a coding-focused data mixture, then post-trained on demonstrations of various coding tasks and tool use as well as demonstrations of correct refusal behaviors based on xAI's safety policy.
86
+
[Registration is required for access to the grok-code-fast-1 model](https://aka.ms/xai/grok-code-fast-1).
89
87
90
88
- Grok 4 Fast, an efficiency-optimized language model that delivers near-Grok 4 reasoning capabilities with significantly lower latency and cost, and can bypass reasoning entirely for ultra-fast applications. It is trained for safe and effective tool use, with built-in refusal behaviors, a fixed safety-enforcing system prompt, and input filters to prevent misuse.
89
+
[Registration is required for access to grok-4-fast-reasoning](https://aka.ms/xai/grok-4-fast-reasoning) and [grok-4-fast-non-reasoning](https://aka.ms/xai/grok-4-fast-non-reasoning).
91
90
92
91
| Model | Type | Capabilities | Deployment type (region availability) | Project type |
Copy file name to clipboardExpand all lines: articles/ai-services/speech-service/includes/release-notes/release-notes-tts.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
@@ -1049,7 +1049,7 @@ For more information, see the [language and voice list](../../language-support.m
1049
1049
* Support 2 new styles `sports-commentary`, `sports-commentary-excited` in public preview for `zh-CN-YunjianNeural`. See [voice styles and roles](../../language-support.md?tabs=tts).
1050
1050
* Support 1 new style `advertisement-upbeat` in public preview for `zh-CN-YunhaoNeural`. See [voice styles and roles](../../language-support.md?tabs=tts).
1051
1051
* The `cheerful` and `sad` styles for `fr-FR-DeniseNeural` are generally available in all regions.
1052
-
* SSML updated to support MathML elements for en-US and en-AU voices. Learn more at [speech synthesis markup](../../speech-synthesis-markup-pronunciation.md#pronunciation-with-mathml).
1052
+
* SSML updated to support MathML elements for en-US and en-AU voices. Learn more at [speech synthesis markup](../../speech-synthesis-markup-pronunciation.md#mathematical-expressions-reading).
![learn-build-service-prod[bot]](https://avatars.githubusercontent.com/in/237442?v=4&size=40){kind=avatar}
0 commit comments