update

Author: laujan

articles/ai-foundry/how-to/develop/cloud-evaluation.md

@@ -243,6 +243,18 @@ print("Versioned evaluator id:", registered_evaluator.id)
 
 After you log your custom evaluator to your Azure AI project, you can view it in your [Evaluator library](../evaluate-generative-ai-app.md#view-and-manage-the-evaluators-in-the-evaluator-library) under the **Evaluation** tab of your Azure AI project.
 
+### Troubleshooting: Job Stuck in Running State
+
+If your evaluation job remains in the **Running** state for an extended period when using Azure AI Foundry Project or Hub, this may be because the Azure OpenAI model you selected does not have enough capacity.
+
+**Resolution**
+
+Cancel the current evaluation job.
+
+Increase the model capacity to handle larger input data.
+
+Re-run the evaluation.
+
 ## Related content
 
 - [Evaluate your generative AI applications locally](./evaluate-sdk.md)

articles/ai-foundry/openai/api-version-lifecycle.md

@@ -6,7 +6,7 @@ manager: nitinme
 ms.service: azure-ai-foundry
 ms.subservice: azure-ai-foundry-openai
 ms.topic: conceptual 
-ms.date: 09/05/2025
+ms.date: 10/01/2025
 author: mrbullwinkle
 ms.author: mbullwin
 recommendations: false
@@ -27,7 +27,7 @@ Previously, Azure OpenAI received monthly updates of new API versions. Taking ad
 
 Starting in August 2025, you can now opt in to our next generation v1 Azure OpenAI APIs which add support for:
 
-- Ongoing access to the latest features with no need specify new `api-version`'s each month.
+- Ongoing access to the latest features with no need to specify new `api-version`'s each month.
 - Faster API release cycle with new features launching more frequently.
 - OpenAI client support with minimal code changes to swap between OpenAI and Azure OpenAI when using key-based authentication.
 - OpenAI client support for token based authentication and automatic token refresh without the need to take a dependency on a separate Azure OpenAI client.
@@ -43,29 +43,13 @@ For the initial v1 Generally Available (GA) API launch we're only supporting a s
 
 ## Code changes
 
-# [API Key](#tab/key)
+# [Python](#tab/python)
 
-### Last generation API 
+### v1 API
 
-```python
-import os
-from openai import AzureOpenAI
-
-client = AzureOpenAI(
-    api_key=os.getenv("AZURE_OPENAI_API_KEY"),  
-    api_version="2025-04-01-preview",
-    azure_endpoint="https://YOUR-RESOURCE-NAME.openai.azure.com")
-    )
-
-response = client.responses.create(
-    model="gpt-4.1-nano", # Replace with your model deployment name 
-    input="This is a test."
-)
+[Python v1 examples](./supported-languages.md)
 
-print(response.model_dump_json(indent=2)) 
-```
-
-### Next generation API
+**API Key**:
 
 ```python
 import os
@@ -88,33 +72,13 @@ print(response.model_dump_json(indent=2))
 - `base_url` passes the Azure OpenAI endpoint and `/openai/v1` is appended to the endpoint address.
 - `api-version` is no longer a required parameter with the v1 GA API.
 
-# [Microsoft Entra ID](#tab/entra)
-
-### Last generation API 
+**API Key** with environment variables set for `OPENAI_BASE_URL` and `OPENAI_API_KEY`:
 
 ```python
-from openai import AzureOpenAI
-from azure.identity import DefaultAzureCredential, get_bearer_token_provider
-
-token_provider = get_bearer_token_provider(
-    DefaultAzureCredential(), "https://cognitiveservices.azure.com/.default"
-)
-
-client = AzureOpenAI(
-  azure_endpoint = "https://YOUR-RESOURCE-NAME.openai.azure.com/", 
-  azure_ad_token_provider=token_provider,
-  api_version="2025-04-01-preview"
-)
-
-response = client.responses.create(
-    model="gpt-4.1-nano", # Replace with your model deployment name 
-    input="This is a test."
-)
-
-print(response.model_dump_json(indent=2)) 
+client = OpenAI()
 ```
 
-### Next generation API
+**Microsoft Entra ID**:
 
 > [!IMPORTANT]
 > Handling automatic token refresh was previously handled through use of the `AzureOpenAI()` client. The v1 API removes this dependency, by adding automatic token refresh support to the `OpenAI()` client.
@@ -143,40 +107,150 @@ print(response.model_dump_json(indent=2))
 - `base_url` passes the Azure OpenAI endpoint and `/openai/v1` is appended to the endpoint address.
 - `api_key` parameter is set to `token_provider`, enabling automatic retrieval and refresh of an authentication token instead of using a static API key.
 
-# [REST](#tab/rest)
+# [C#](#tab/dotnet)
+
+### v1 API
 
-### Last generation API 
+[C# v1 examples](./supported-languages.md)
 
 **API Key**:
 
-```bash
-curl -X POST https://YOUR-RESOURCE-NAME.openai.azure.com/openai/responses?api-version=2025-04-01-preview \
-  -H "Content-Type: application/json" \
-  -H "api-key: $AZURE_OPENAI_API_KEY" \
-  -d '{
-     "model": "gpt-4.1-nano",
-     "input": "This is a test"
-    }'
+```csharp
+OpenAIClient client = new(
+    new ApiKeyCredential("{your-api-key}"),
+    new OpenAIClientOptions()
+    {
+        Endpoint = new("https://YOUR-RESOURCE-NAME.openai.azure.com/openai/v1/"),
+    })
 ```
 
 **Microsoft Entra ID**:
 
-```bash
-curl -X POST https://YOUR-RESOURCE-NAME.openai.azure.com/openai/responses?api-version=2025-04-01-preview \
-  -H "Content-Type: application/json" \
-  -H "Authorization: Bearer $AZURE_OPENAI_AUTH_TOKEN" \
-  -d '{
-     "model": "gpt-4.1-nano",
-     "input": "This is a test"
-    }'
+```csharp
+#pragma warning disable OPENAI001
+
+BearerTokenPolicy tokenPolicy = new(
+    new DefaultAzureCredential(),
+    "https://cognitiveservices.azure.com/.default");
+OpenAIClient client = new(
+    authenticationPolicy: tokenPolicy,
+    options: new OpenAIClientOptions()
+    {
+        Endpoint = new("https://YOUR-RESOURCE-NAME.openai.azure.com/openai/v1/"),
+    })
+```
+
+# [JavaScript](#tab/javascript)
+
+### v1 API
+
+[JavaScript v1 examples](./supported-languages.md)
+
+**API Key**:
+
+```javascript
+const client = new OpenAI({
+    baseURL: "https://YOUR-RESOURCE-NAME.openai.azure.com/openai/v1/",
+    apiKey: "{your-api-key}" 
+});
+```
+
+**API Key** with environment variables set for `OPENAI_BASE_URL` and `OPENAI_API_KEY`:
+
+```javascript
+const client = new OpenAI();
+```
+
+**Microsoft Entra ID**:
+
+```javascript
+const tokenProvider = getBearerTokenProvider(
+    new DefaultAzureCredential(),
+    'https://cognitiveservices.azure.com/.default');
+const client = new OpenAI({
+    baseURL: "https://YOUR-RESOURCE-NAME.openai.azure.com/openai/v1/",
+    apiKey: tokenProvider
+});
+```
+
+# [Go](#tab/go)
+
+### v1 API
+
+[Go v1 examples](./supported-languages.md)
+
+**API Key**:
+
+```go
+client := openai.NewClient(
+    option.WithBaseURL("https://YOUR-RESOURCE-NAME.openai.azure.com/openai/v1/"),
+    option.WithAPIKey("{your-api-key}")
+)
+```
+
+**API Key** with environment variables set for `OPENAI_BASE_URL` and `OPENAI_API_KEY`:
+
+```go
+client := openai.NewClient()
+```
+
+
+**Microsoft Entra ID**:
+
+```go
+tokenCredential, err := azidentity.NewDefaultAzureCredential(nil)
+
+client := openai.NewClient(
+    option.WithBaseURL("https://YOUR-RESOURCE-NAME.openai.azure.com/openai/v1/"),
+    azure.WithTokenCredential(tokenCredential)
+)
 ```
 
-### Next generation API
+# [Java](#tab/Java)
+
+[Java v1 examples](./supported-languages.md)
+
+### v1 API
+
+**API Key**:
+
+```java
+
+OpenAIClient client = OpenAIOkHttpClient.builder()
+                .baseUrl("https://YOUR-RESOURCE-NAME.openai.azure.com/openai/v1/")
+                .apiKey(apiKey)
+                .build();
+```
+
+**API Key** with environment variables set for `OPENAI_BASE_URL` and `OPENAI_API_KEY`:
+
+```java
+OpenAIClient client = OpenAIOkHttpClient.builder()
+                .fromEnv()
+                .build();
+```
+
+**Microsoft Entra ID**:
+
+```java
+Credential tokenCredential = BearerTokenCredential.create(
+        AuthenticationUtil.getBearerTokenSupplier(
+                new DefaultAzureCredentialBuilder().build(),
+                "https://cognitiveservices.azure.com/.default"));
+OpenAIClient client = OpenAIOkHttpClient.builder()
+        .baseUrl("https://YOUR-RESOURCE-NAME.openai.azure.com/openai/v1/")
+        .credential(tokenCredential)
+        .build();
+```
+
+# [REST](#tab/rest)
+
+### v1 API
 
 **API Key**:
 
 ```bash
-curl -X POST https://YOUR-RESOURCE-NAME.openai.azure.com/openai/v1/responses?api-version=preview \
+curl -X POST https://YOUR-RESOURCE-NAME.openai.azure.com/openai/v1/responses \
   -H "Content-Type: application/json" \
   -H "api-key: $AZURE_OPENAI_API_KEY" \
   -d '{
@@ -188,7 +262,7 @@ curl -X POST https://YOUR-RESOURCE-NAME.openai.azure.com/openai/v1/responses?api
 **Microsoft Entra ID**:
 
 ```bash
-curl -X POST https://YOUR-RESOURCE-NAME.openai.azure.com/openai/v1/responses?api-version=preview \
+curl -X POST https://YOUR-RESOURCE-NAME.openai.azure.com/openai/v1/responses \
   -H "Content-Type: application/json" \
   -H "Authorization: Bearer $AZURE_OPENAI_AUTH_TOKEN" \
   -d '{

articles/ai-foundry/openai/azure-government.md

@@ -56,6 +56,27 @@ To request quota increases for these models, submit a request at [https://aka.ms
 
 <br>
 
+### Model Retirements
+In some cases, models are retired in Azure Governmen ahead of dates in the commercial cloud. General information on model retirement policies, dates, and other details can be found at [Azure OpenAI in Azure AI Foundry model deprecations and retirements](/azure/ai-foundry/openai/concepts/model-retirements). The following shows model retirement differences in Azure Government. 
+
+| Model                     | Version			      | Azure Government Status	  | Public Retirement date             |
+| --------------------------|-------------------|:--------------------------|------------------------------------|
+| `gpt-35-turbo`            | 1106				      | Retired                   | November 11, 2025                  |
+| `gpt-4`                   | turbo-2024-04-09	| Retired                   | November 11, 2025                  |
+
+<br>
+
+### Deafault Model Versions 
+In some cases, new model versions are designated as default in Azure Governmen ahead of dates in the commercial cloud. General information on model upgrades can be found at [Working with Azure OpenAI models](/azure/ai-foundry/openai/how-to/working-with-models?tabs=powershell&branch=main#model-deployment-upgrade-configuration)
+
+The following shows default model differences in Azure Government. 
+
+| Model     | Azure Government Default Version | Public Default Version | Default upgrade date          |
+|-----------|----------------------------------|------------------------|-------------------------------|
+|  `gpt-4o` | 2024-11-20                       | 2024-08-06             | Starting on October 13, 2025  |
+
+<br>
+
 ## Azure OpenAI features
 
 The following feature differences exist when comparing Azure OpenAI in Azure Government vs commercial cloud.

articles/ai-foundry/openai/how-to/provisioned-throughput-onboarding.md

@@ -83,14 +83,14 @@ For example, for gpt-5 1 output token counts as 8 input tokens towards your util
 > [!NOTE]
 > gpt-4.1, gpt-4.1-mini and gpt-4.1-nano don't support long context (requests estimated at larger than 128k prompt tokens).
 
-|Topic| **gpt-5** | **gpt-4.1** | **gpt-4.1-mini** | **gpt-4.1-nano** | **o3** | **o4-mini** | 
-| --- |  --- | --- |  --- |  --- | --- |  --- |
-|Global & data zone provisioned minimum deployment| 15  | 15|15| 15 | 15 | 15 |
-|Global & data zone provisioned scale increment| 5 | 5|5| 5 | 5 | 5 |
-|Regional provisioned minimum deployment| 50 | 50|25| 25 |50 |25|
-|Regional provisioned scale increment| 50 | 50|25| 25 | 50 | 25|
-|Input TPM per PTU| 4,750 | 3,000|14,900| 59,400 | 3,000 | 5,400 |
-|Latency Target Value| 99% > 50 Tokens Per Second\* | 99% > 80 Tokens Per Second\* | 99% > 90 Tokens Per Second\*| 99% > 100 Tokens Per Second\* | 99% > 80 Tokens Per Second\* | 99% > 90 Tokens Per Second\* |
+|Topic| **gpt-5** | **gpt-5-mini** | **gpt-4.1** | **gpt-4.1-mini** | **gpt-4.1-nano** | **o3** | **o4-mini** | 
+| --- | --- |  --- | --- |  --- |  --- | --- |  --- |
+|Global & data zone provisioned minimum deployment| 15 | 15  | 15|15| 15 | 15 | 15 |
+|Global & data zone provisioned scale increment| 5 | 5 | 5|5| 5 | 5 | 5 |
+|Regional provisioned minimum deployment| 50 | 25 | 50|25| 25 |50 |25|
+|Regional provisioned scale increment| 50 | 25 | 50|25| 25 | 50 | 25|
+|Input TPM per PTU| 4,750  | 23,750 | 3,000|14,900| 59,400 | 3,000 | 5,400 |
+|Latency Target Value| 99% > 50 Tokens Per Second\* | 99% > 80 Tokens Per Second\* | 99% > 80 Tokens Per Second\* | 99% > 90 Tokens Per Second\*| 99% > 100 Tokens Per Second\* | 99% > 80 Tokens Per Second\* | 99% > 90 Tokens Per Second\* |
 
 \* Calculated as p50 request latency on a per 5 minute basis.
 

articles/ai-foundry/openai/how-to/spillover-traffic-management.md

@@ -6,15 +6,15 @@ ms.author: mopeakande
 ms.service: azure-ai-foundry
 ms.subservice: azure-ai-foundry-openai
 ms.topic: how-to
-ms.date: 09/03/2025
+ms.date: 10/02/2025
 ---
 
 # Manage traffic with spillover for provisioned deployments
 
 Spillover manages traffic fluctuations on provisioned deployments by routing overage traffic to a corresponding standard deployment. Spillover is an optional capability that can be set for all requests on a given deployment or can be managed on a per-request basis. When spillover is enabled, Azure OpenAI in Azure AI Foundry Models sends any overage traffic from your provisioned deployment to a standard deployment for processing.
 
 > [!NOTE]
-> Spillover is currently not available for the `/v1` [API endpoint](../reference-preview-latest.md) or [responses API](./responses.md).
+> Spillover is currently not available for the [responses API](./responses.md).
 
 ## Prerequisites
 - You need to have a provisioned managed deployment and a standard deployment.
@@ -27,7 +27,28 @@ Spillover manages traffic fluctuations on provisioned deployments by routing ove
 To maximize the utilization of your provisioned deployment, you can enable spillover for all global and data zone provisioned deployments. With spillover, bursts or fluctuations in traffic can be automatically managed by the service. This capability reduces the risk of experiencing disruptions when a provisioned deployment is fully utilized. Alternatively, spillover is configurable per-request to provide flexibility across different scenarios and workloads. Spillover can also now be used for the [Azure AI Foundry Agent Service](../../agents/overview.md).  
 
 ## When does spillover come into effect?
-When spillover is enabled for a deployment or configured for a given inference request, spillover is initiated when a non-200 response code is received for a given inference request. When a request results in a non-200 response code, the Azure OpenAI automatically sends the request from your provisioned deployment to your standard deployment to be processed. Even if a subset of requests is routed to the standard deployment, the service prioritizes sending requests to the provisioned deployment before sending any overage requests to the standard deployment, which may incur additional latency.
+When you enable spillover for a deployment or configure it for a given inference request, spillover initiates when a specific non-`200` response code is received for a given inference request as a result of one of these scenarios:
+
+- Provisioned throughput units (PTU) are completely used, resulting in a `429` response code.
+
+- You send a long context token request, resulting in a `400` error code. For example, when using `gpt 4.1` series models, PTU supports only context lengths less than 128k and returns HTTP 400.
+
+- Server errors when processing your request, resulting in error code `500` or `503`.
+
+When a request results in one of these non-`200` response codes, Azure OpenAI automatically sends the request from your provisioned deployment to your standard deployment to be processed.
+
+> [!NOTE]
+> Even if a subset of requests is routed to the standard deployment, the service prioritizes sending requests to the provisioned deployment before sending any overage requests to the standard deployment, which might incur additional latency.
+
+## How to know a request spilled over
+
+The following HTTP response headers indicate that a specific request spilled over:
+
+- `x-ms-spillover-from-<deployment-name>`. This header contains the PTU deployment name. The presence of this header indicates that the request was a spillover request.
+
+- `x-ms-<deployment-name>`. This header contains the name of the deployment that served the request. If the request spilled over, the deployment name is the name of the standard deployment.
+
+For a request that spilled over, if the standard deployment request failed for any reason, the original PTU response is used in the response to the customer. The customer sees a header `x-ms-spillover-error` that contains the response code of the spillover request (such as `429` or `500`) so that they know the reason for the failed spillover.
 
 ## How does spillover affect cost?
 Since spillover uses a combination of provisioned and standard deployments to manage traffic fluctuations, billing for spillover involves two components:
@@ -124,4 +145,4 @@ Applying the `IsSpillover` split lets you view the requests to your deployment t
 ## See also
 
 * [What is provisioned throughput](../concepts/provisioned-throughput.md)
-* [Onboarding to provisioned throughput](./provisioned-throughput-onboarding.md)
+* [Onboarding to provisioned throughput](./provisioned-throughput-onboarding.md)