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/foundry-models/how-to/quickstart-github-models.md
+31-23Lines changed: 31 additions & 23 deletions
Display the source diff
Display the rich diff
Original file line number
Diff line number
Diff line change
@@ -15,75 +15,83 @@ recommendations: false
15
15
16
16
# Upgrade from GitHub Models to Azure AI Foundry Models
17
17
18
-
In this article, you learn to develop a generative AI application by starting from GitHub Models and then upgrading your experience by deploying an Azure AI Services resource with Azure AI Foundry Models.
18
+
In this article, you learn to develop a generative AI application by starting from GitHub Models and then upgrade your experience by deploying an Azure AI Services resource with Azure AI Foundry Models.
19
19
20
-
[GitHub Models](https://docs.github.com/en/github-models/) are useful when you want to find and experiment with AI models for free as you develop a generative AI application. When you're ready to bring your application to production, upgrade your experience by deploying an Azure AI Services resource in an Azure subscription and start using Azure AI Foundry Models service. You don't need to change anything else in your code.
20
+
[GitHub Models](https://docs.github.com/en/github-models/) are useful when you want to find and experiment with AI models for free as you develop a generative AI application. When you're ready to bring your application to production, upgrade your experience by deploying an Azure AI Services resource in an Azure subscription and start using Foundry Models. You don't need to change anything else in your code.
21
21
22
22
The playground and free API usage for GitHub Models are [rate limited](https://docs.github.com/en/github-models/prototyping-with-ai-models#rate-limits) by requests per minute, requests per day, tokens per request, and concurrent requests. If you get rate limited, you need to wait for the rate limit that you hit to reset before you can make more requests.
23
23
24
24
## Prerequisites
25
25
26
26
To complete this tutorial, you need:
27
27
28
-
* A GitHub account with access to [GitHub Models](https://docs.github.com/en/github-models/).
29
-
* An Azure subscription. If you don't have one, you're prompted to create or update your Azure account to a Standard account when you're ready to deploy your model to production.
28
+
- A GitHub account with access to [GitHub Models](https://docs.github.com/en/github-models/).
29
+
- An Azure subscription with a valid payment method. If you don't have an Azure subscription, create a [paid Azure account](https://azure.microsoft.com/pricing/purchase-options/pay-as-you-go) to begin. Alternatively, you can wait until you're ready to deploy your model to production, at which point you'll be prompted to create or update your Azure account to a standard account.
30
+
-[Foundry Models from partners and community](../concepts/models-from-partners.md) require access to **Azure Marketplace**. Ensure you have the [permissions required to subscribe to model offerings](configure-marketplace.md). [Foundry Models sold directly by Azure](../concepts/models-sold-directly-by-azure.md) don't have this requirement.
31
+
30
32
31
33
## Upgrade to Azure AI Foundry Models
32
34
33
35
The rate limits for the playground and free API usage help you experiment with models and develop your AI application. When you're ready to bring your application to production, use a key and endpoint from a paid Azure account. You don't need to change anything else in your code.
34
36
35
37
To get the key and endpoint:
36
38
37
-
1. Go to [GitHub Models](https://github.com/marketplace/models) and select the model you're interested in.
39
+
1. Go to [GitHub Models](https://github.com/marketplace/models) and select a model to land on its playground. This article uses Mistral Large 24.11.
38
40
39
-
1.In the playground for your model, select **Get API key**.
41
+
1.Type in some prompts or use some of the suggested prompts to interact with the model in the playground.
40
42
41
-
1. Select **Get production key**.
43
+
1. Select **Use this model** from the playground. This action opens up a window to "Get started with Models in your codebase".
42
44
43
-
:::image type="content" source="../media/quickstart-github-models/github-models-upgrade.gif" alt-text="An animation showing how to upgrade GitHub Models to get a production ready resource." lightbox="../media/quickstart-github-models/github-models-upgrade.gif":::
45
+
1. In the "Configure authentication" step, select **Get Azure AI key** from the "Azure AI" section.
44
46
45
-
1. If you don't have an Azure account, select **Create my account** and follow the steps to create one.
47
+
:::image type="content" source="../media/quickstart-github-models/github-models-get-production-key.png" alt-text="A screenshot showing how to get the Azure AI production key from the playground of a GitHub Model." lightbox="../media/quickstart-github-models/github-models-get-production-key.png":::
46
48
47
-
1. If youhave an Azure account, select **Sign back in**.
49
+
1. If you're already signed in to your Azure account, skip this step. However, if you don't have an Azure account or you're not signed in to your account, follow these steps:
48
50
49
-
1. If your existing account is a free account, you first have to upgrade to a Standard plan. Once you upgrade, go back to the playground and select **Get API key**again, then sign in with your upgraded account.
51
+
1. If you don't have an Azure account, select **Create my account**and follow the steps to create one.
50
52
51
-
1.When you sign in to your Azure account, you're taken to [Azure AI Foundry > GitHub](https://ai.azure.com/GitHub). It might take one or two minutes to load your initial model details in AI Foundry.
53
+
1.Alternatively, if you have an Azure account, select **Sign back in**. If your existing account is a free account, you first have to upgrade to a standard plan.
52
54
53
-
1. The page loads with your model's details. Select the**Deploy** button to deploy the model to your account.
55
+
1. Return to the model's playground and select**Get Azure AI key** again.
54
56
55
-
1. When it's deployed, your model's API Key and endpoint appear in the Overview. Use these values in your code to use the model in your production environment.
57
+
1. Sign in to your Azure account.
58
+
59
+
1. You're taken to [Azure AI Foundry > GitHub](https://ai.azure.com/GitHub), and the page loads with your model's details. It might take one or two minutes to load your model details in Azure AI Foundry.
56
60
57
-
At this point, the model you selected is ready to consume.
61
+
1. For [Foundry Models from partners and community](../concepts/models-from-partners.md), you need to subscribe to Azure Marketplace. This requirement applies to Mistral-Large-2411, for example. Select **Agree and Proceed** to accept the terms.
58
62
59
-
## Upgrade your code to use the new endpoint
63
+
1. Select the **Deploy** button to deploy the model to your account.
60
64
61
-
After you configure your Azure AI Services resource, you can start using it from your code. To use the Azure AI Services resource, you need the endpoint URL and key, which you can find in the **Overview** section:
65
+
1. When your deployment is ready, you land on your project's **Overview** page, where you can see the Azure AI Foundry project's endpoint.
62
66
63
-
:::image type="content" source="../media/overview/overview-endpoint-and-key.png" alt-text="Screenshot showing how to get the URL and key associated with the resource." lightbox="../media/overview/overview-endpoint-and-key.png":::
67
+
1. To get the specific model's endpoint URL and API key, go to the **Models + endpoints** tab in the left pane of the Azure AI Foundry portal and select the deployed model. The endpoint's target URI and API key are visible on the deployment's details page. Use these values in your code to use the model in your production environment.
64
68
65
-
You can use any of the supported SDKs to get predictions from the endpoint. The following SDKs are officially supported:
69
+
:::image type="content" source="../media/quickstart-github-models/deployment-endpoint-and-key.png" alt-text="Screenshot showing how to get the URL and key associated with the deployment." lightbox="../media/quickstart-github-models/deployment-endpoint-and-key.png":::
70
+
71
+
## Use the new endpoint
72
+
73
+
To use your deployed model with code, you need the model's endpoint URL and key, which you saw in the previous section. You can use any of the supported SDKs to get predictions from the endpoint. The following SDKs are officially supported:
66
74
67
75
* OpenAI SDK
68
76
* Azure OpenAI SDK
69
77
* Azure AI Inference SDK
70
78
71
-
For more details and examples, see the [supported languages and SDKs](../supported-languages.md) section. The following example shows how to use the Azure AI Foundry Models SDK with the newly deployed model:
79
+
For more details and examples, see [supported languages and SDKs](../supported-languages.md). The following example shows how to use the Azure AI Inference SDK with the newly deployed model:
Use the parameter `model="<deployment-name>` to route your request to this deployment. *Deployments work as an alias of a given model under certain configurations*. See [Routing](inference.md#routing) concept page to learn how Azure AI Services route deployments.
87
+
Use the parameter `model="<deployment-name>` to route your request to this deployment. *Deployments work as an alias of a given model under certain configurations*.
80
88
81
89
> [!IMPORTANT]
82
90
> Unlike GitHub Models where all the models are already configured, the Azure AI Services resource allows you to control which models are available in your endpoint and under which configuration. Add as many models as you plan to use before indicating them in the `model` parameter. Learn how to [add more models](../../model-inference/how-to/create-model-deployments.md) to your resource.
83
91
84
92
## Explore additional features
85
93
86
-
Azure AI Foundry Models supports additional features that aren't available in GitHub Models, including:
94
+
Azure AI Foundry Models supports extra features that aren't available in GitHub Models, including:
87
95
88
96
*[Explore the model catalog](https://ai.azure.com/github/models) to see more models.
Copy file name to clipboardExpand all lines: articles/ai-foundry/foundry-models/includes/create-model-deployments/intro.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
@@ -21,4 +21,4 @@ To complete this article, you need:
21
21
* An Azure AI Foundry project. This kind of project is managed under an Azure AI Foundry resource (formerly known as Azure AI Services resource). If you don't have a Foundry project, see [Create a project for Azure AI Foundry (Foundry projects)](../../../how-to/create-projects.md).
22
22
23
23
24
-
*[Models from Partners and Community](../../concepts/models.md#models-from-partners-and-community) require access to **Azure Marketplace**. Ensure you have the [permissions required to subscribe to model offerings](../../how-to/configure-marketplace.md). [Models Sold Directly by Azure](../../concepts/models.md#models-sold-directly-by-azure) don't have this requirement.
24
+
*[Foundry Models from partners and community](../../concepts/models-from-partners.md) require access to **Azure Marketplace**. Ensure you have the [permissions required to subscribe to model offerings](../../how-to/configure-marketplace.md). [Foundry Models sold directly by Azure](../../concepts/models-sold-directly-by-azure.md) don't have this requirement.
Copy file name to clipboardExpand all lines: articles/ai-foundry/how-to/continuous-evaluation-agents.md
+17-3Lines changed: 17 additions & 3 deletions
Display the source diff
Display the rich diff
Original file line number
Diff line number
Diff line change
@@ -16,9 +16,7 @@ author: lgayhardt
16
16
17
17
Continuous evaluation for Agents provides near real-time observability and monitoring for your AI application. Once enabled, this feature continuously evaluates agent interactions at a set sampling rate to provide insights into quality, safety, and performance with metrics surfaced in the Foundry Observability dashboard. By using continuous evaluation, you're able to identify and troubleshoot issues early, optimize agent performance, and maintain safety. Evaluations are also connected to [traces](./develop/trace-application.md) to enable detailed debugging and root cause analysis.
@@ -31,6 +29,22 @@ Continuous evaluation for Agents provides near real-time observability and monit
31
29
2. Select **Monitoring** on the left-hand menu and go to **Application Analytics**.
32
30
3. Connect your Application Insights resource to the project.
33
31
32
+
## Azure AI Foundry project configuration and region support
33
+
34
+
Since the evaluators use hosted evaluation LLMs in the Azure AI Foundry evaluation service, they require your Azure AI project information to be instantiated. The Azure AI project must be in a supported region:
![learn-build-service-prod[bot]](https://avatars.githubusercontent.com/in/237442?v=4&size=40){kind=avatar}
0 commit comments