---
title: Azure AI Foundry
description: >-
  Use the Azure AI Foundry integration to track the usage and performance of
  your model deployments.
breadcrumbs: Docs > Integrations > Azure AI Foundry
---

# Azure AI Foundry
Integration version1.0.0
## Overview{% #overview %}

AI Foundry is a platform for designing, customizing, and managing AI applications. Use the Azure AI Foundry integration to track the usage and performance of your model deployments.

## Setup{% #setup %}

### Installation{% #installation %}

If you haven't already, set up the [Microsoft Azure integration](https://docs.datadoghq.com/integrations/azure.md) first. There are no other installation steps.

## Data Collected{% #data-collected %}

### Metrics{% #metrics %}

|  |
|  |
| **azure.cognitiveservices\_accounts.model\_availability\_rate**(gauge)                          | Availability percentage with the following calculation: (total calls - server errors)/total calls. Server errors include any 500-level HTTP responses.*Shown as percent*                 |
| **azure.cognitiveservices\_accounts.model\_requests**(count)                                    | Number of calls made to the model API over a period of time. Applies to PTU, PTU-managed, and pay-as-you-go deployments.                                                                 |
| **azure.cognitiveservices\_accounts.input\_tokens**(count)                                      | Number of prompt tokens processed (input) on a model. Applies to PTU, PTU-managed, and pay-as-you-go deployments.                                                                        |
| **azure.cognitiveservices\_accounts.output\_tokens**(count)                                     | Number of tokens generated (output) from an OpenAI model. Applies to PTU, PTU-managed, and pay-as-you-go deployments.                                                                    |
| **azure.cognitiveservices\_accounts.total\_tokens**(count)                                      | Number of inference tokens processed on a model. Calculated as prompt tokens (input) plus generated tokens (output). Applies to PTU, PTU-managed, and pay-as-you-go deployments.         |
| **azure.cognitiveservices\_accounts.count**(gauge)                                              | Count of CognitiveServices accounts.                                                                                                                                                     |
| **azure.cognitiveservices\_accounts.annotated\_pages**(count)                                   | Total number of pages processed with annotations. Applies to PTU, ptu-managed and pay-as-you-go deployments.                                                                             |
| **azure.cognitiveservices\_accounts.audio\_input\_tokens**(count)                               | Number of audio prompt tokens processed (input) on an openai model. Applies to ptu-managed model deployments.                                                                            |
| **azure.cognitiveservices\_accounts.audio\_output\_tokens**(count)                              | Number of audio prompt tokens generated (output) on an openai model. Applies to ptu-managed model deployments.                                                                           |
| **azure.cognitiveservices\_accounts.generated\_images**(count)                                  | Total number of images generated. Applies to PTU, ptu-managed and pay-as-you-go deployments.                                                                                             |
| **azure.cognitiveservices\_accounts.total\_pages**(count)                                       | Total number of pages processed. Applies to PTU, ptu-managed and pay-as-you-go deployments.                                                                                              |
| **azure.cognitiveservices\_accounts.realtime\_api\_seconds\_used**(count)                       | RealtimeAPI number of seconds used.                                                                                                                                                      |
| **azure.cognitiveservices\_accounts.prompt\_tokens\_read\_from\_cache**(count)                  | Total number of tokens read from the cache. Applies to anthropic model deployments. Surfaced in response usage section as cache_read_input_tokens.                                       |
| **azure.cognitiveservices\_accounts.prompt\_tokens\_written\_to\_cache\_1\_hour\_ttl**(count)   | Number of prompt tokens used to create the 1 hour entry. Applies to anthropic model deployments. Surfaced in response usage section as cache_creation.ephemeral_1h_input_tokens.         |
| **azure.cognitiveservices\_accounts.prompt\_tokens\_written\_to\_cache\_5\_minute\_ttl**(count) | Number of prompt tokens used to create the 5 minute cache entry. Applies to anthropic model deployments. Surfaced in response usage section as cache_creation.ephemeral_5m_input_tokens. |
| **azure.cognitiveservices\_accounts.voice\_live\_audio\_input\_tokens**(count)                  | Number of audio input tokens, excluding cached tokens.                                                                                                                                   |
| **azure.cognitiveservices\_accounts.voice\_live\_audio\_output\_tokens**(count)                 | Number of audio output tokens.                                                                                                                                                           |
| **azure.cognitiveservices\_accounts.voice\_live\_cached\_audio\_input\_tokens**(count)          | Number of cached audio input tokens.                                                                                                                                                     |
| **azure.cognitiveservices\_accounts.voice\_live\_cached\_text\_input\_tokens**(count)           | Number of cached text input tokens.                                                                                                                                                      |
| **azure.cognitiveservices\_accounts.voice\_live\_text\_input\_tokens**(count)                   | Number of text input tokens, excluding cached tokens.                                                                                                                                    |
| **azure.cognitiveservices\_accounts.voice\_live\_text\_output\_tokens**(count)                  | Number of text output tokens.                                                                                                                                                            |

### Service Checks{% #service-checks %}

Azure AI Foundry does not include any service checks.

### Events{% #events %}

Azure AI Foundry does not include any events.

## Troubleshooting{% #troubleshooting %}

Need help? Contact [Datadog support](https://docs.datadoghq.com/help/).