Lenovo expands hybrid AI portfolio with inferencing platforms and agentic AI tools

Lenovo has expanded its Hybrid AI Advantage portfolio with a range of AI inferencing platforms and agentic AI tools aimed at helping enterprises deploy AI agents faster and reduce token costs across on-premise, hybrid and cloud environments.

The announcements centre on two new Hybrid AI Platform configurations and a set of one-click agentic AI deployment capabilities, with Lenovo positioning the portfolio as a way to address what it describes as unsustainable costs in enterprise AI rollouts.

The new CPU-only Lenovo Hybrid AI Platform, built with Red Hat AI Enterprise and powered by Intel Xeon 6 processors, is designed to process roughly twice as many AI requests concurrently, according to Lenovo. The platform targets workloads including retrieval-augmented generation, HR support and customer service, and combines Lenovo servers, storage and Red Hat AI Enterprise into a validated architecture for private AI deployment.

A second platform - the Lenovo Hybrid AI Platform (221) - is available in two configurations: one built on Canonical Ubuntu and Kubernetes for development speed and data sovereignty, and one on Red Hat AI Enterprise for governed production workloads. Lenovo said systems can be ready in as little as a few weeks.

The company claimed its inference-optimised architecture can deliver up to eight times lower cost per token than cloud infrastructure-as-a-service environments, and up to 18 times lower cost per million tokens compared with model-as-a-service APIs - though those figures apply to workloads requiring sustained CPU and GPU utilisation and are Lenovo's own comparisons.

On the agentic AI side, Lenovo said it is introducing one-click deployment of autonomous and long-running agents across devices and data centre infrastructure.

New use cases include knowledge synthesis agents that Lenovo said can save thousands of employee hours, according to independent analysis. The company is also co-developing NVIDIA NemoClaw skills for AI operations use cases, and building agentic AI blueprints for the Lenovo ThinkStation PGX workstation.

Trust and governance capabilities announced alongside the inferencing and agentic tools include a Nutanix Compute Only Cluster on Lenovo ThinkSystem servers, the Lenovo XClarity One unified management platform, and supply chain hardware root-of-trust protections.

"Enterprises are moving beyond AI experimentation and demanding measurable business outcomes," said Ashley Gorakhpurwalla, president of infrastructure solutions at Lenovo.

"Lenovo is uniquely positioned to help customers deploy AI where it creates the greatest value - while dramatically improving token economics, accelerating time-to-value, and maintaining the trust, security, and governance organisations require."

The announcements extend Lenovo's existing Hybrid AI Advantage programme, which spans AI PCs, workstations, servers and cloud-connected infrastructure. Ecosystem partners named across the new platforms include NVIDIA, Intel, Red Hat and Canonical.