Microsoft is heralding the arrival of an official OpenAI library for .NET developers. The OpenAI library supports the complete OpenAI API and OpenAI’s latest flagship model, GPT-4o, which can reason across audio, vision, and text in real time.

Announced June 6 and available now in a first beta, the OpenAI .NET API library is accessible from NuGet.

Resulting from Microsoft’s collaboration with OpenAI, the OpenAI .NET API library provides sync and async access to the OpenAI REST API from .NET applications. It supports capabilities including Assistants v2, for building AI assistants within applications, and Chat Completions, for taking a list of messages as input and returning a model message. The library includes extensibility that enables the community to build libraries on top and access to streaming completions via IAsyncEnumerable<T>.

The OpenAI .NET API library is supported on GitHub and will be kept up to date with the latest features from OpenAI, Microsoft said. Work will continue in coming months to gather feedback, improve the library, and offer a stable NuGet package. The OpenAI .NET API library is compatible with .NET Standard 2.0 applications. However, some code examples in the OpenAI .NET API library document might depend on newer language features. Developers will need an API key to call the OpenAI REST API.

Microsoft last month announced investments to expand the AI ecosystem for .NET developers. AI attention included end-to-end scenarios for building AI-enabled applications, embracing the AI ecosystem, and integration with cloud services.