Lenovo has announced new hybrid Ai solutions built with Nvidia technology that it says will help organisations improve productivity and efficiency by accelerating deployment of agentic Ai.
Unveiled at Nvidia GTC on March 25, the new offerings aim to address a range of enterprise use cases through a full-stack, validated approach that combines Lenovo’s hybrid infrastructure with the latest from Nvidia’s Blackwell platform.
Lenovo Chairman and CEO Yuanqing Yang said the partnership brings together Ai models, data and computing across devices, edge and cloud environments.
“Lenovo Hybrid AI Advantage with Nvidia integrates services and Blackwell-accelerated infrastructure to help enterprises scale agentic AI,” Yang said. “This enables access to public and private models that improve efficiency, security and customisation.”
Ai Deployment Still Challenging
While many organisations are eager to deploy Ai, Lenovo cited new IDC research showing return on investment remains a major obstacle. The report found Ai spending has tripled, but business leaders are still cautious.
Lenovo is positioning its hybrid Ai approach as a way to simplify deployment and improve ROI by delivering agentic Ai that can perform multistep planning, code generation and reasoning.
“Organisations often struggle with fragmented AI adoption,” a Lenovo spokesperson said. “With this offering, we are simplifying that process, enabling Ai to work across devices, edge and cloud.”
The solutions use Nvidia’s RTX PRO 6000 Blackwell Server Edition GPUs and new networking components to support a range of applications such as real-time video generation and agentic Ai.
Use Cases Validated Internally
Lenovo said its new platform has already been tested internally, showing measurable gains in productivity and automation.
One example is content generation, where the company claims to have achieved up to an eightfold increase in speed. Customer service processes saw a 50 per cent improvement in efficiency, while internal knowledge assistants improved legal team productivity by 80 per cent and accuracy by 45 per cent.
The Lenovo Ai Library forms part of the solution stack and includes a set of tested use cases that can be quickly customised by enterprises. These span personal devices, data centres and public cloud environments.
The Lenovo Ai Knowledge Assistant was also showcased at the event, featuring a digital human interface designed to help attendees navigate GTC. It was built using Lenovo’s agentic Ai platform and Nvidia’s AI Blueprint and NIM microservices.
Infrastructure Designed For Scale
The company is promoting its hybrid Ai factory model as a key enabler for organisations looking to scale Ai deployments. This includes workstations, servers, storage and software built using Lenovo Validated Designs and Nvidia reference architectures.
Key components include Lenovo ThinkSystem SR675, 680 and 685 servers, powered by Nvidia Hopper and Blackwell GPUs, as well as Nvidia Grace CPUs and BlueField DPUs. These platforms are intended to support both Ai training and inference workloads across a modular deployment model.
One example is the ThinkSystem SR675 V3 server, which can operate as a standalone unit or scale to rack-level deployments. It uses Nvidia H200 NVL GPUs and the Nvidia Ai Enterprise software stack to manage high-performance computing and Ai workloads.
Other offerings include liquid-cooled infrastructure via Lenovo Neptune, AI-ready ThinkStation PX workstations with Nvidia Blackwell GPUs, and a Fast Start service that promises business value within 90 days.
Sector Adoption Expands
Lenovo said hybrid Ai is already having an impact in several industries including healthcare, finance and manufacturing. In Germany, the company is working with the Technical University of Darmstadt on a project using the Lenovo ThinkSystem SC777 V4 Neptune system to support the Lichtenberg NHR-Stage 1 supercomputer.
The deployment features Lenovo’s sixth generation direct water-cooling platform and aims to support scientific research using the next-generation Grace-Blackwell architecture.
In healthcare, software firm AISHA has built a model that analyses full-body MRI scans in 30 minutes. Built on Lenovo and Nvidia’s hybrid Ai infrastructure, the system delivers insights more than 99 per cent faster than traditional methods.
“Without the power of the Lenovo and Nvidia solution, the model would simply not be able to exist,” said Dr Juan Pablo Reyes Gonzalez, Head of AISHA. “Lenovo and Nvidia are unmatched in the field of AI.”
Looking Ahead
With 44 per cent of businesses in Australia and New Zealand planning to adopt Ai over the next year, Lenovo is aiming to position its offerings as a lower-risk way to deliver value.
Jensen Huang, Nvidia founder and CEO, said the collaboration focuses on real-world deployment.
“AI agents that can reason and adapt are redefining how we work,” Huang said. “Nvidia and Lenovo are providing the infrastructure to bring these capabilities to enterprise environments.”
More information is available at: https://www.lenovo.com/us/en/servers-storage/solutions/ai