When Qualcomm formally introduced itsSnapdragon X Eliteplatforms atComputexearlier this month, the company almost solely focused on the AI capabilities of these system-on-chips and only briefly touched upon CPU and GPU. But to take advantage of Qualcomm’s Hexagon neural processing unit (NPU), one needs AI software that can use it.
Today, Qualcomm developed, optimized, and validated its AI models for its 45 TOPSHexagonNPU. The models are available at the company’s AI Hub to registered software developers who can use already trained models to build their artificial intelligence-enabled applications.
The Qualcomm AI Hub contains various AI models optimized for deployment on edge devices powered by Qualcomm’sSnapdragon X Eliteprocessors, including image classification, object detection, semantic segmentation, and generative AI. Also, the hub provides tools and documentation for deploying AI models in applications such as vision,audio, and speech easily to make development as smooth as possible.
To further simplify things, developers can filter models by tags such as ‘backbone, ‘foundation,’ ‘quantized,’ and ‘real-time.’ The site also includes detailed examples and code snippets for deploying and running AI models on devices based on the Snapdragon X Elite processor family.
By making its trained AI models optimized for 45 TOPS Hexagon NPU, Qualcomm ensures that over time, hopefully within a year, there will be a host of AI-enabled applications that can take advantage of its Snapdragon X Elite, which will catalyze demand for PCs based on Qualcomm’s SoCs.
In fact, part of the reason Qualcomm focused its Snapdragon X Elite announcement on AI and not on its premium CPU and GPU is that it wanted to attract software developers’ attention to what it thinks is the killer feature of 2024 PCs.
Get Tom’s Hardware’s best news and in-depth reviews, straight to your inbox.
In addition, by making its models optimized for particular hardware available, Qualcomm also ensures that AI programs will run best on its Snapdragon X Elite platforms, which will give it a competitive edge against AMD and Intel, who are also about to roll out their new CPUs with high-performance NPUs for AI PCs.
Anton Shilov is a contributing writer at Tom’s Hardware. Over the past couple of decades, he has covered everything from CPUs and GPUs to supercomputers and from modern process technologies and latest fab tools to high-tech industry trends.
