Intel has announced the availability of its 12th Gen Intel® Core™ SoC processors for IoT Edge. Representing a new lineup of purpose-built edge products optimized for Internet of Things (IoT) applications, this first-of-its-kind socketed system-on-chip (SoC) delivers high performance integrated graphics and media processing for visual compute workloads, a compact footprint to enable smaller innovative form factor designs, and a wide operating thermal design power (TDP) that enables fanless designs and helps customers achieve product sustainability goals.

“As the digitization of business processes continues to accelerate — fueled by workforce demand, supply chain constraints and changing consumer behavior — the amount of data created at the edge and the need for it to be processed and analyzed locally continues to explode. Intel understands the challenges that businesses face — across a wide range of vertical industries — and is committed to help them continue to deliver innovative use cases.” –Jeni Panhorst, Intel vice president and general manager of the Network and Edge Compute Division

Why It Matters: Digital transformation at the edge requires increased processing power and AI inference performance to future-proof AI workloads. 12th Gen Intel Core SoC processors for IoT Edge respond to these performance demands and expand implementation configurability and overall solution flexibility to enable original equipment manufacturers and original design manufacturers to quickly integrate – and ship – solutions that fit the wide variety of unique vertical market and edge-specific use cases. Additionally, the SoC processors feature manageability capabilities from top to bottom, including Intel vPro options for best-in-class remote control and manageability essential for managing and servicing systems deployed at the IoT edge.

About 12th Gen Intel Core SoC processors for IoT Edge: Delivering up to 4 times faster graphics¹ and up to 6.6 times faster GPU image classification inference performance² compared with 10th Gen Intel® Core™ desktop processors in a 12W to 65W design, 12th Gen Intel Core SoC processors for IOT Edge include Intel® Thread Director, which intelligently directs the operating system to assign the right workload to the right core. With up to 14 cores and 20 threads, the SoC processors reach up to 1.32 times faster single-thread performance³ and up to 1.27 times faster multithread performance⁴ compared with 10th Gen Intel® Core™ desktop processors.

12th Gen Intel Core SoC processors for IoT Edge support high-performing AI for inferencing and machine vision. Up to 96 graphics execution units allow for a high degree of parallelization in AI workloads, while built-in AI acceleration on the CPU from Intel® Deep Learning Boost (Intel® DL Boost) provides additional inferencing performance. These processors fully support Intel® Distribution of OpenVINO™ toolkit optimizations and cross-architecture inferencing.

“By combining the CPU with the highest performing integrated graphics, enhanced visual compute, and AI into one compact processing configuration, 12th Gen Intel Core SoC processors for IoT Edge make it easier for customers to unlock new opportunities from customized retail point-of-sale solutions to precise imaging and pattern recognition for improved healthcare diagnoses,” Panhorst said.

12th Generation Intel Core processors for IoT Edge deliver benefits across industries:

Retail, banking, hospitality and education customers will be able to expand remote control and manageability capabilities of systems at the edge, drive value from frictionless point-of-sale and quickly respond to changing supply and demand.
Industrial manufacturing customers can take better advantage of industrial PCs, edge servers, advanced controllers, machine vision systems and virtualized control platforms.
Healthcare customers will be able to deliver enhanced ultrasound imaging, medical carts, endoscopy and clinical devices at the edge.

Performance varies by use, configuration, and other factors. Learn more at Intel.com/PerformanceIndex.

¹Up to 4 times faster graphics performance as measured by 3DMark - Fire Strike - Graphics Score

²Up to 6.6 times faster in GPU image classification inference performance MLPerf v1.1 ResNet50 - Offline - int8 – GPU. MLPerf™ Inference Edge v1.1 Inference ResNet-v1.5; Result not verified by the MLCommons™ Association. The MLPerf name and logo are trademarks of MLCommons Association in the United States and other countries. All rights reserved. Unauthorized use strictly prohibited. See www.mlcommons.org for more information.

³Up to 1.32 times faster in single-thread performance as measured by SPECrate2017_int_base (1-copy) using InteI Compiler version 2021.2

⁴Up to 1.27 times faster in multi-thread performance as measured by SPECrate2017_int_base (n-copy) using InteI Compiler version 2021.2