UL Solutions Unveils Inaugural Showcase of UL Procyon® AI Inference Benchmark for Windows at Snapdragon Summit

UL Solutions announced that its UL Procyon® Artificial Intelligence (AI) Inference Benchmark for Windows demonstrated the AI inference performance of the Snapdragon® X Elite Platform during the 2023 Snapdragon Summit.

By Brian K. Neal

UL Solutions, a global leader in applied safety science, has unveiled the results of its UL Procyon® Artificial Intelligence (AI) Inference Benchmark for Windows, showcasing the AI inference capabilities of the Snapdragon® X Elite Platform during the 2023 Snapdragon Summit.

An inference engine, comprised of hardware and software, makes decisions based on stored rules and information, and its performance can vary depending on the device and other factors.

The UL Procyon AI Inference Benchmark for Windows compares various inference engines from different vendors, evaluating the machine learning inference performance of Windows operating system devices using common machine vision models. This benchmark provides valuable insights for technology companies, enabling them to assess inference performance and compare devices for AI use cases.

Jukka Makinen, Managing Director in the Software and Advisory group at UL Solutions, commented, “With the unprecedented access to machine learning and artificial intelligence tools that consumers now have in their daily lives, technology vendors are utilizing performance insights to optimize inference engines for better performance with hardware and specific use cases. Our UL Procyon AI Inference Benchmark for Windows empowers the technology industry with a performance measurement tool and insights that can assist in the development of faster AI-driven products and features.”

The AI workloads in the UL Procyon AI Inference Benchmark for Windows involve standard machine vision tasks such as image classification, image segmentation, object detection, and super-resolution. These tasks are executed using various neural networks and run on the device’s central processing unit (CPU), graphics processing unit (GPU), or a dedicated AI accelerator, enabling a comprehensive comparison of hardware performance.

Surya Chittuluri, Director of Product Management at Qualcomm Technologies, Inc., acknowledged the significance of the insights gained from the benchmark, stating, “The insights gained from the UL Procyon AI Inference Benchmark for Windows enabled Qualcomm Technologies to optimize the performance of our machine learning tools and present data to highlight the new capabilities of our leading Snapdragon X Elite Platform. We are pleased to be among the industry partners working with UL Solutions in the ongoing development of its AI benchmark in support of faster machine learning.”