Microsoft teams up with Intel to boost AI on Windows with DirectML and NPUs

Microsoft and Intel are introducing the Neural Processing Unit (NPU) powered by DirectML on Windows. In easier words, Microsoft has partnered with Intel to make Windows computers smarter when handling tasks involving artificial intelligence (AI). This collaboration focuses on enhancing the capabilities of Windows devices using NPUs (Neural Processing Units).

What is DirectML?

DirectML, also known as Direct Machine Learning, is a powerful tool for developers that empowers them to leverage the capabilities of modern GPUs to accelerate machine learning (ML) tasks.

What are NPUs?

Neural Processing Units (NPUs) are specialized processors designed to accelerate machine learning algorithms. Think of them as turbocharged CPUs optimized for the specific needs of AI.

NPUs are specifically designed for machine learning tasks, offering significant performance advantages over GPUs for certain workloads.

DirectML’s integration with NPUs expands the range of hardware options available for AI development on Windows. This opens up opportunities for developers working with different budgets and needs.

The early availability of DirectML with NPU support in developer preview next year will allow developers to start experimenting and preparing for the broader rollout later in 2024.

DirectML’s low-level API and integration with popular frameworks like ONNX Runtime and Olive simplify the building and deploying AI models on Windows, hence acting like a powerful toolkit for AI development on Windows.

The announcement mentions further collaboration with key partners and expanded capabilities for DirectML with NPUs in the future.

Overall, Microsoft and Intel are teaming up to make it easier for software developers to use these NPUs effectively within Windows computers.

More info here.