First Official OpenAI Library for .NET Goes Beta -- Visual Studio Magazine

First Official OpenAI Library for .NET Goes Beta

By David Ramel
06/07/2024

Although it seems Microsoft and OpenAI have been deeply intertwined partners for a long time, they are only now getting around to releasing an official OpenAI library for .NET developers, joining existing community libraries.

The first beta of the effort is an OpenAI NuGet package supporting .NET 6 and .NET Standard 2.0, now at version 2.0.0-beta.2 and listing some 868,000 downloads.

**[Click on image for larger view.]** OpenAI .NET API Library *(source: OpenAI).*

The new library was foreshadowed at Microsoft's recent Build 2024 developer conference, when the company announced:

One of our most exciting collaborations this year has been with OpenAI. We partnered with them to deliver an official .NET library, which is set to be released later this month. This collaboration and new SDK ensures that .NET developers have a delightful experience and will have parity with other programming language libraries that you may be familiar with. It also provides support for the latest OpenAI features and models, such as GPT4o and Assistants v2, and a unified experience across OpenAI and Azure OpenAI.

That's cutting-edge tech, as GPT-4o (with the "o" standing for omni) is the latest/greatest flagship large language model (LLM) from OpenAI, while the Assistants API lets developers create helper AI constructs that can respond to instructions and leverage models, tools and files to respond to user queries. Another supported feature, the Chat Completions API, can help developers build apps that generate natural language responses to user inputs. Independent, actionable helpers that communicate via natural language is a hot area of AI development right now.

In announcing OpenAI's library yesterday (June 6), Microsoft's .NET development team said it's based on the OpenAPI specification, providing features -- in addition to those listed above -- such as:

Extensibility to enable the community to build libraries on top
Sync and async APIs for ease of use and efficiency
Access to streaming completions via IAsyncEnumerable<T>

Joining other official OpenAI libraries for Python and TypeScript/JavaScript developers, Microsoft said the .NET library will ensure smooth and supported integration with OpenAI and Azure OpenAI for developers in its camp.

As far as unsupported integration, the .NET team recognized pioneering work by Roger Pincombe and the existing community options Betalgo.OpenAI by Betalgo and OpenAI-DotNet by RageAgainstThePixel.

"The .NET library is developed and supported on GitHub and will be kept up to date with the latest features from OpenAI," the team said yesterday. "Work will continue over the next few months to gather feedback to improve the library and release a stable NuGet package."

About the Author

David Ramel is an editor and writer at Converge 360.

Printable Format

comments powered by Disqus

Featured

Windows Community Toolkit v8.2 Adds Native AOT Support

Microsoft shipped Windows Community Toolkit v8.2, an incremental update to the open-source collection of helper functions and other resources designed to simplify the development of Windows applications. The main new feature is support for native ahead-of-time (AOT) compilation.
New 'Visual Studio Hub' 1-Stop-Shop for GitHub Copilot Resources, More

Unsurprisingly, GitHub Copilot resources are front-and-center in Microsoft's new Visual Studio Hub, a one-stop-shop for all things concerning your favorite IDE.
Mastering Blazor Authentication and Authorization

At the Visual Studio Live! @ Microsoft HQ developer conference set for August, Rockford Lhotka will explain the ins and outs of authentication across Blazor Server, WebAssembly, and .NET MAUI Hybrid apps, and show how to use identity and claims to customize application behavior through fine-grained authorization.
Linear Support Vector Regression from Scratch Using C# with Evolutionary Training

Dr. James McCaffrey from Microsoft Research presents a complete end-to-end demonstration of the linear support vector regression (linear SVR) technique, where the goal is to predict a single numeric value. A linear SVR model uses an unusual error/loss function and cannot be trained using standard simple techniques, and so evolutionary optimization training is used.
Low-Code Report Says AI Will Enhance, Not Replace DIY Dev Tools

Along with replacing software developers and possibly killing humanity, advanced AI is seen by many as a death knell for the do-it-yourself, low-code/no-code tooling industry, but a new report belies that notion.