UL Procyon® AI Inference Benchmark for Windows Released 22nd March 2023

Today, March 22, 2023, UL Solutions released The UL Procyon AI Inference Benchmark for Windows. Simple and easy to run, the benchmark compares multiple inference engines from different vendors, measuring the machine learning inference performance of Windows devices using common machine vision models.  

Addressing the demand for faster machine learning performance

Machine learning applications are rapidly growing as it becomes more accessible to integrate and deploy artificial intelligence (AI) tools into everyday applications. With the demand for faster machine learning performance, major hardware vendors have been optimizing their inference engines to provide the best possible performance on their hardware.  

Hundreds of publications use benchmarks developed by UL Solutions to test and review the latest PC hardware and mobile devices. Our UL Procyon benchmarking suite is closely developed with our industry partners to ensure that every Procyon benchmark is accurate, relevant and impartial.  

We designed the UL Procyon® AI Inference Benchmark for Windows for industry vendors developing AI-specific hardware who need relevant, standardized tools for assessing AI inference performance. Providing insights into how AI inference engines perform on your hardware in a Windows environment, the benchmark monitors both CPU and GPU load, helping you decide which engines to support to achieve optimal performance.  

Relevant AI workloads

Like the UL Procyon AI Inference benchmark on Android, the AI workloads used in the UL Procyon AI Inference Benchmark for Windows are standard machine vision tasks, such as image classification, image segmentation, object detection and super-resolution.  

These tasks are executed using various well- known, state-of-the-art neural networks. They can run on the device’s CPU, GPU or a dedicated AI accelerator for comparing hardware performance differences. Software development kits (SDKs) used to measure the AI interference performance include Microsoft® Windows ML, Qualcomm ® SNPE, Intel® OpenVINO™ and NVIDIA® TensorRT™.  

UL Procyon AI Inference Benchmark for Windows

The UL Procyon AI Inference Benchmark for Windows integrates commonly available frameworks, SDKs and application programming interfaces (APIs) for performing high-speed inferences across various hardware platforms. UL Solutions will continue development after its release to maintain the benchmark’s industry relevance. Potential future updates include tests for newly developed use cases for AI models or updating AI runtimes to the latest versions.  

The UL Procyon benchmark suite offers flexible licensing, letting you choose the benchmarks that best meet your needs. You can buy just one benchmark or add more in any combination.  

Find out more about UL Procyon benchmarks.