HomeWinBuzzer NewsMicrosoft Unveils DirectML Developer Preview with NPU Support for Intel Core Ultra...

Microsoft Unveils DirectML Developer Preview with NPU Support for Intel Core Ultra CPUs

Microsoft's DirectML now supports Intel's NPU in Core Ultra CPUs, boosting AI tasks on PCs.


In a significant advancement for , has announced the release of the developer preview version of its DirectML API, which now includes support for Neural Processor Units (NPUs), particularly those embedded in Intel's Core Ultra CPUs. The preview, referred to as DirectML 1.13.1, alongside ONNX Runtime 1.17, aims to leverage the NPU's capabilities in accelerating AI-driven tasks on compatible personal computers.

AI Enhancement on Intel Core Ultra CPUs

The integration of NPU support within the DirectML framework is a response to the increasing demand for quick and efficient AI processing on local devices. AI applications, such as facial recognition and object classification, stand to benefit substantially from this development. Enhanced by Intel's AI Boost technology, these applications can run more effectively by offloading certain tasks to the dedicated hardware of the NPU.

In a significant showcase of collaboration between industry giants, Intel has highlighted 's incorporation of both the updated DirectML API and Intel's AI technologies in the Samsung Galaxy Book 4 notebook, equipped with Core Ultra chips. Intel asserts that using the DirectML API, Samsung has optimized the AI features within its Gallery Windows app, allowing for a more streamlined and expedited development process.

Preview Limitations and Developer Feedback

The unveiling of this preview remains a work in progress, a point that Microsoft acknowledges. In particular, developers are encouraged to submit feedback and report any issues they encounter via the DirectML repository. This collaborative approach is expected to fine-tune the technology, ensuring the DirectML API can fully harness the potential of NPUs in upcoming software advancements.

Microsoft's update marks a pivotal step in the evolution of AI and machine learning capabilities in mainstream computing devices. As the software and hardware continue to synergize, users can anticipate further performance enhancements and innovative application developments in AI technology.

Luke Jones
Luke Jones
Luke has been writing about all things tech for more than five years. He is following Microsoft closely to bring you the latest news about Windows, Office, Azure, Skype, HoloLens and all the rest of their products.

Recent News