Lenovo Unveils New Solutions for Enterprise Agent AI

Lenovo has announced an expansion of its Lenovo Hybrid AI Advantage solutions portfolio, which includes new platforms designed for the implementation of artificial intelligence and agent AI in businesses, in collaboration with Canonical, Red Hat, NVIDIA, and Nutanix.

Lenovo's New Offerings for Agent AI

The Lenovo Hybrid AI Platform with Red Hat consists of servers with Intel Xeon 6 processors equipped with Red Hat AI Enterprise and is designed for CPU-based inference. Lenovo claims it is ideal for applications such as retrieval-augmented generation, human resources support, and customer assistance. According to Lenovo, the new platform allows for higher throughput, lower latency, and reduced first-token processing times.

The Lenovo Hybrid AI Platform is available in two variants: the first utilizes solutions produced by Canonical, namely Ubuntu and Kubernetes, while the second targets large enterprises and is based on the Red Hat Enterprise ecosystem. These two platforms symbolize Lenovo's strategy, which includes a hybrid approach to AI with processing on-premises or on owned infrastructure, as well as in the cloud, bringing it closer to users, data, and business processes. Lenovo states that this approach can reduce the cost per token by up to 8 times compared to cloud IaaS environments and up to 18 times per million tokens compared to direct purchase.

On the agent AI front, Lenovo has introduced the ability to deploy autonomous AI agents with a “one-click” capability on its platform, allowing customers to quickly experiment with agent AI from workstations (such as ThinkStation PGX or ThinkStation PX) to data centers. For example, Lenovo is developing solutions for IT activities based on NVIDIA NemoClaw, as well as “personal AI factories” for ThinkStation PGX that enable the creation of customized AI agents.

"The data speaks for itself: as highlighted in the Lenovo CIO Playbook 2026, 94% of companies are ready to increase AI investments in the next year. We are seeing companies move beyond mere experimentation to demand concrete and measurable business results," commented Ashley Gorakhpurwalla, President of Infrastructure Solutions at Lenovo.

"In this scenario, Lenovo is in a privileged position to help customers implement AI precisely where it can create the most value. Our mission is to dramatically improve token economics, accelerate time-to-value, while ensuring the trust, security, and governance that organizations require."

"For many CIOs in Europe and the Middle East, the challenge of AI is no longer technological but economic. Companies have already demonstrated the value of AI through pilot projects, but extending it throughout the organization requires a very different approach in terms of costs, governance, and data control," stated Per Overgaard, Executive Director, Europe and the Middle East Infrastructure Solutions Group at Lenovo.

"The focus is shifting from maximizing the number of tokens to maximizing outcomes, as IT managers face increasing pressure to demonstrate the measurable business value of every AI investment. As organizations implement more inference and agent AI workloads, they need the flexibility to run AI exactly where data is created and decisions are made. Lenovo's latest innovations help customers implement AI closer to their data, users, and business processes, improving performance, control, and cost efficiency within the hybrid AI operating model.