Lenovo launches expanded AI inferencing portfolio

Lenovo has unveiled a line of ThinkSystem and ThinkEdge servers designed specifically for AI inferencing workloads. The solutions and software stacks are designed to cater to AI inferencing workloads of any size and are built on Lenovo’s Hybrid AI Advantage platform.

The inferencing-optimised portfolio includes flagship solutions built to run full large language models anywhere for large workloads and accelerated simulation, as well as compact offerings designed to run in retail, telco and industrial environments and in a range of climate conditions. They take advantage of Lenovo Neptune’s air and liquid cooling capability, and use the Lenovo TruScale pay-as-you-go pricing model.

Lenovo’s enterprise AI hardware is the foundation of the Lenovo Hybrid AI Factory platform, a modular framework for building and operating AI solutions at scale. The range has been expanded to support hybrid AI inferencing using solutions including Lenovo ThinkAgile HX and Nutanix AI, Red Hat AI, and Canonical Ubuntu Pro.

In addition, the platform now supports AI inferencing services aimed at optimising server performance, as well as providing ongoing assistance for data centre management.

Futurum estimates that the global AI inference infrastructure market will grow at a 46.3% CAGR over the six years from 2024 to reach US$48.8 billion ($72.9 billion) by 2030.

“Enterprises today need AI that can turn massive amounts of data into insight the moment it’s created,” said Lenovo EVP and President of Infrastructure Solutions Ashley Gorakhpurwalla. “With Lenovo’s new inferencing-optimised infrastructure, we are giving customers that real-time advantage — transforming massive amounts of data into instant, actionable intelligence that fuels stronger decisions, greater security and faster innovation.”

Image credit: iStock.com/monsitj