---
title: Azure AI
description: "Introduction to the Statsig Azure OpenAI integration for logging AI calls, capturing metrics, and running experiments on prompts and model parameters."
product: general
lang: en
token_estimate: 364
---
# Azure AI

Statsig offers SDKs for integrating Azure AI models into server applications. These SDKs simplify the implementation of features like completions and embeddings in your server application. They provide easy-to-use APIs and automatically track metrics such as latency, token length, and model details, which you can use for optimization and experimentation. Use cases include:

> For AI agents: a documentation index is available at [/llms.txt](/llms.txt). Append `.md` to any page URL for markdown, or send `Accept: text/markdown`.

- Implement Azure AI Models in your code with a [single lightweight framework](https://docs.statsig.com/integrations/azureai/model-client)
- [Stream completions](https://docs.statsig.com/integrations/azureai/completions) and [generate embeddings](https://docs.statsig.com/integrations/azureai/embeddings)
- [Capture invocation and usage metrics](https://docs.statsig.com/integrations/azureai/capturing-metrics) with no extra work
- [Run A/B tests on parameters](https://docs.statsig.com/integrations/azureai/running-experiments) like model, prompt, temperature and more

![Azure AI integration architecture diagram](https://docs.statsig.com/images/integrations/azureai/introduction/b23c79c3-8501-4390-a3f3-3496970eb272.png)

## Supported SDKs

- Node JS: https://github.com/statsig-io/azureai-nodejs/
- Python: https://github.com/statsig-io/azureai-python/
- .Net: https://github.com/statsig-io/azureai-dotnet/