Intel Unveils Next-Gen Edge AI Processors

Intel has made a significant announcement regarding its latest lineup of edge AI processors, poised to revolutionize AI-enabled devices across various vertical industries such as retail, manufacturing, healthcare, and aerospace.

Presented at the Embedded World conference, these cutting-edge processors, developed in collaboration with Altera, an Intel company, encompass a range of FPGA processors alongside Atom and Core CPUs, and Arc discrete graphics.

A distinctive feature of Intel’s new edge AI processors lies in their integration with other Intel products, seamlessly connected through Intel’s OneAPI library.

Dan Rodriguez, Corporate Vice President and General Manager of Intel’s Network and Edge Solutions Group, highlighted the transformative potential of these processors, stating, “This next generation of Intel edge-optimized processors and discrete GPUs unleashes powerful AI capabilities to help businesses more seamlessly incorporate AI alongside compute, media, and graphics workloads.”

Among the highlights of the new offerings is the availability of Agilex 5 SoC FPGAs, targeting mid-level devices. Intel boasts up to twice the performance per watt compared to competing 7-nanometer FPGAs, designed to deliver high performance with lower power consumption within a modern SoC subsystem, featuring small form factor package options.

Additionally, Intel has made Agilex 5 development kits widely accessible, complemented by the FPGA AI Suite. This suite empowers developers to leverage existing AI frameworks, alongside Intel’s OpenVINO toolkit, facilitating the creation and integration of AI intellectual property blocks into FPGA designs.

In tandem with FPGA developments, Intel has introduced Intel Core Ultra Processors for Edge, equipped with Intel Arc GPU and neural processing units (NPU), boasting up to five times better image classification inference performance compared to previous iterations. These processors target applications in retail, education, smart cities, and industrial sectors, supporting generative AI and graphics workloads at the edge.

Furthermore, Intel has expanded its portfolio with lower-power Core processors for edge applications, delivering up to 2.57 times greater graphics performance compared to predecessors.

While Intel’s Atom processors have faced challenges in competing with Arm in the embedded space, the x7000C series continues to thrive, tailored for enterprise networking and telecommunications devices. These processors offer packet processing and deep learning inference capabilities, catering to industrial and manufacturing applications through the Atom x7000RE Series, featuring built-in deep learning inference capabilities and supporting fanless designs for AI-driven automation applications.