据IT之家报道，微软近日宣布，将在明年初为DirectML机器学习框架添加NPU支持，从而适配英特尔酷睿Ultra等内置NPU的芯片。这意味着，开发者可以通过跨平台推理引擎“ONNX Runtime”等API，使用芯片内置的NPU运行AI模型，从而改善模型运行效能。
据了解，NPU（NVIDIA Processing Unit）是NVIDIA推出的一种专门为AI计算而设计的GPU。与传统的GPU相比，NPU拥有更高的性能和能效，可大大缩短训练和推理的时间。
此次微软加入NPU支持，将为AI开发者提供更多选择，同时也将推动整个AI行业的发展。
英文标题：Microsoft to add NPU support to DirectX in Q1 of next year
英文翻译：Microsoft to add NPU support to DirectX in Q1 of next year.
新闻正文：
Microsoft has recently announced that it will add NPU support to the DirectML machine learning framework in the first quarter of next year. This will enable the framework to be used with chips that have NPU integrated, such as Intel Core i9-11900K. This move is seen as a significant development in the field of AI, as it provides developers with more options and also boosts the overall growth of the AI industry.
据了解，NPU（NVIDIA Processing Unit）是NVIDIA推出的一种专门为AI计算而设计的GPU。与传统的GPU相比，NPU拥有更高的性能和能效，可大大缩短训练和推理的时间。
The addition of NPU support to DirectML will provide developers with more options and resources for training and inference of AI models. It is expected that this move will drive further growth in the AI industry and accelerate the development of AI technology.
新闻来源：
IT之家，2022年12月11日
【来源】https://www.ithome.com/0/740/014.htm
Views: 1