Lenovo's official site for press materials and original stories about the vision and passion behind the technology.

With the launch of Lenovo AI Express, organizations have a faster path to validated, right-sized AI infrastructure.

The Lenovo ThinkAgile VX850 V4 is a 2U 4-socket system that features Intel Xeon 6 processors. These systems are based on the ThinkSystem SR850 V4 with up to 86 cores per processor, up to 16TB of TruDDR5 memory with up to 6400 MHz memory speeds, and an extensive selection of internal storage and networking options. VMware offers a unique, software-defined approach to hyperconvergence, leveraging the hypervisor to deliver compute, storage and management in a tightly integrated software stack. This product guide provides essential pre-sales information to understand the VX systems, their key features and specifications, components and options, and configuration guidelines. This guide is intended for technical specialists, sales specialists, sales engineers, IT architects, and other IT professionals who want to learn more about the VX systems and consider their use in IT solutions.

The collaboration builds on a broader commitment to advancing AI security through openness and industry-wide cooperation.

Nutanix Agentic AI combines Nutanix Enterprise AI, Nutanix Kubernetes Platform, and Nutanix Cloud Infrastructure to run and govern agentic applications across private data centers and public clouds. Lenovo ThinkAgile HX V4 and FX V4 systems with Intel Xeon 6 processors provide the integrated on-premises infrastructure for this stack. Intel testing of Nutanix Enterprise AI 2.5 on a Lenovo ThinkSystem SR650 V4 showed up to 2x the output throughput of the prior generation, and Lenovo testing on ThinkAgile HX650 V4 kept Llama 3 8B second-token latency below 100 ms through batch size 8. This brief describes the architecture, use cases, and a CPU-first path to private enterprise AI.

VMware Cloud Foundation (VCF) 9 on Lenovo ThinkAgile VX V3/V4 and FX V4 hyperconverged systems, Lenovo ThinkSystem V3/V4 servers, and Lenovo ThinkSystem Storage delivers a cloud operating model inside the enterprise data center. The combined solution unifies compute, storage, networking, automation, and lifecycle management in one software-defined stack, supporting traditional applications, Kubernetes services, and private AI on shared infrastructure. Organizations gain public cloud agility with the cost predictability, data sovereignty, and governance of on-premises infrastructure.

Lenovo Edge solutions, including the ThinkAgile MX455 V3 and the 1U and 2U rack-mounted ThinkAgile MX V3 and V4 platforms, are well suited for deployment in remote locations, providing customers with a compact, cost-effective hybrid cloud infrastructure solution. This document describes how to deploy a single-node Azure Local instance on Lenovo ThinkAgile MX systems, using either the Azure portal or an Azure Resource Manager (ARM) template. It is intended for IT personnel who are familiar with Windows Server and hybrid cloud solutions.

Partnership gives a practical pathway to measure emissions, build decarbonization roadmaps, and demonstrate sustainability progress.

Lenovo enhances virtualization portfolio with new solutions and services designed to help enterprises modernize with ease.

Lenovo has achieved the highest-possible rating in the 2026 Sustainability Ratings released by Hang Seng Indexes Company, advancing from its AA+ rating last year to AAA this year.

Enterprises are shifting customer, IoT, and AI-driven workloads onto MongoDB's flexible document model to keep pace with fast-changing data. Lenovo ThinkAgile HX650 V4 and FX650 V4 hyperconverged systems, built on Intel Xeon 6 processors and the Nutanix Cloud Platform, give MongoDB a validated, high-performance foundation for private cloud, analytics, and AI/ML initiatives. In Yahoo Cloud Serving Benchmark (YCSB) benchmark testing, a standalone MongoDB instance running on a four-node ThinkAgile HX650 V4 Nutanix cluster sustained peak throughput of 116,149 operations per second with sub-millisecond latency on read-intensive workloads, showing that the platform delivers the throughput and consistency modern MongoDB deployments require.

Eighth annual Best Workplaces for Innovators list recognizes businesses that are creating cultures where innovation can thrive at every level.

After years of vendors heralding the arrival of fine-tuned models to save enterprises from the latency, unpredictable costs, and strenuous upgrade lifecycles of cloud-based model APIs, the moment for fine-tuned models has finally arrived. While frontier models are massive, general-purpose, and must be run from cloud servers, fine-tuned models (mostly composed of post-trained open-weight models) […]

Delray Beach, FL, Sept. 03, 2026 (GLOBE NEWSWIRE) -- According to MarketsandMarkets™, the Data Center Solutions Market is projected to grow from USD 535.45 billion in 2026 to USD 1,336.55 billion by 2031, registering a CAGR of 20.1% during the forecast period. Growth is being fueled by demand for resilient, high-density computing environments supported by intelligent power systems, advanced cooling, software-defined infrastructure, and automation platforms. Get PDF Brochure — Market Data & Segme

Lenovo ThinkAgile FX V4 Series is a multi-vendor hyperconverged appliance: a single factory-integrated system, built on Lenovo ThinkSystem servers with Intel Xeon 6 processors, that runs the VMware, Nutanix, or Microsoft hyperconverged software stack. When an organization's platform strategy changes, Lenovo Professional Services converts the existing FX V4 hardware to the new stack rather than replacing it. Factory software preload, Lenovo Best Recipes, integrated Lenovo XClarity lifecycle management, and Lenovo Premier Support give infrastructure teams appliance-grade simplicity while keeping the software decision reversible.

Lenovo ThinkAgile FX V4 Series is a multi-vendor hyperconverged appliance: a single factory-integrated system, built on Lenovo ThinkSystem servers with Intel Xeon 6 processors, that runs the VMware, Nutanix, or Microsoft hyperconverged software stack. When an organization's platform strategy changes, Lenovo Professional Services converts the existing FX V4 hardware to the new stack rather than replacing it. Factory software preload, Lenovo Best Recipes, integrated Lenovo XClarity lifecycle management, and Lenovo Premier Support give infrastructure teams appliance-grade simplicity while keeping the software decision reversible.

As GPU refresh cycles shrink toward 18 to 24 months, Dynamic Lifecycle Innovations’ Josh Humm says data center operators need to plan for decommissioning, data security and residual...

Overview According to DataM Intelligence the global software defined data center SDDC market size is estimated at approximately USD 84 27 billion in 2026 derived from the report s stated market value and 23 40 CAGR The market is forecast ...

A site-level outage puts business-critical applications and the VMware Cloud Foundation management stack at risk at the same moment, and manual recovery procedures make that moment longer and less predictable. This solution pairs VMware Live Site Recovery with Lenovo ThinkAgile VX to deliver orchestrated failover, non-disruptive recovery testing, and automated failback between a protected site and a recovery site. Because Lenovo separates the recovery outcome from a single infrastructure shape, the recovery site can be built on ThinkAgile VX with vSAN ESA, on Lenovo ThinkSystem external storage, or on a combination of both, so recovery economics follow business requirements instead of mirroring production by default.

VMware Explore 2026 field report: Lenovo on turnkey AI factories, the RAM crunch, enterprise liquid cooling, and what comes next.

Recover business-critical workloads with confidence, validate, cleanse, and restore inside an isolated recovery "clean room" that keeps your data on infrastructure you control.

VMware Cloud Foundation 9.1 Memory Tiering, validated on Lenovo ThinkAgile VX V4 with Intel Xeon 6 processors, removes that tradeoff by pairing high speed DRAM with low cost, high capacity NVMe SSDs such as Micron 7600 SSDs to transparently double usable memory. Organizations can run twice the virtual machines per node on the same hardware, reduce total cost of acquisition by up to 45 percent, and retain more than 90 percent of performance across enterprise workloads.

Organizations don’t organically mature into effective AI adoption. At Forrester’s CX Forum East, Colleen Fazio explained how companies can build a solid human foundation for AI.

U.S. data center vacancy hit a record low of 1.4% in H1 2026 as AI-driven demand absorbed supply faster than it was built.

Lenovo ThinkAgile MX delivers a validated hyperconverged platform for Microsoft hybrid cloud workloads, and Microsoft Azure Arc extends the Azure control plane to that infrastructure so on-premises and edge resources can be governed with the same tools as resources running in Azure. This implementation guide walks through the process end to end using a two-node ThinkAgile MX650 V3 Azure Local cluster: onboarding servers and virtual machines to Azure Arc, deploying Azure Local from the Azure Arc portal, and Arc-enabling virtual machines created both inside and outside the portal. It then covers day-to-day operations from the Azure Arc portal, including monitoring and security through Azure Monitor and Microsoft Defender for Cloud, cluster and node health views, Arc-enabled Kubernetes clusters, and SQL Server inventory. Preview capabilities for Machine Provisioning and Site Management are also introduced. This paper is for infrastructure architects and IT administrators who are deploying or operating ThinkAgile MX in a hybrid environment.

Enterprise infrastructure no longer sits in one place. Workloads run in the datacenter, at branch and edge sites, and across multiple public clouds, and each environment tends to bring its own management tooling. This brief describes how Lenovo ThinkAgile MX, running Microsoft Azure Local and connected through Azure Arc, gives IT organizations a single Azure-native control plane for servers, virtual machines, Kubernetes clusters, and SQL Server instances that remain on-premises or at the edge.

This technical brief outlines an approach to building a modern private cloud using VMware Cloud Foundation (VCF) on Lenovo ThinkSystem servers, ThinkAgile VX/FX platforms, and ThinkSystem storage systems. It addresses the shift toward private cloud strategies driven by cost predictability, data sovereignty, and the need to support data-intensive workloads such as analytics and AI. The document describes how VCF integrates compute, storage, networking, and lifecycle management to deliver a consistent cloud operating model within enterprise data centers. It also highlights deployment options across hyperconverged and disaggregated architectures using Lenovo infrastructure. The role of external shared storage alongside HCI is discussed in the context of scaling, performance, and workload diversity. The combined solution supports multiple protocols and workload types, including virtual machines and containers. Emphasis is placed on enabling efficient resource utilization, predictable performance, and operational control. The architecture supports incremental expansion and alignment with hybrid cloud strategies and focuses on practical design considerations for implementing a flexible and scalable private cloud platform.

Lenovo ThinkAgile VX and FX V4 with VMware Cloud Foundation deliver a validated, factory-integrated platform for running Oracle Database 23ai on an all-NVMe, software-defined foundation. In testing, a single four-node cluster sustained 36.8 million transactions per minute across eight Oracle database VMs, pairing mission-critical performance with hyperconverged simplicity and single-vendor support.

Dell’Oro forecasts Data Center Physical Infrastructure revenue to reach $120 billion by 2030, driven by AI capacity expansion. Learn more

Microsoft SQL Server 2025 delivers a modern, enterprise-grade relational data platform for mission-critical OLTP, analytics, and AI workloads. Deployed on Lenovo ThinkAgile HX V4 and FX V4 hyperconverged systems with Intel Xeon 6 Processors and Nutanix Cloud Platform, organizations gain dramatically simplified infrastructure operations and a single unified foundation for database, analytics, and enterprise AI initiatives. In Lenovo validated testing, a single ThinkAgile HX650 V4 node configured with Intel Xeon 6 64-core processors sustained up to 11.42 million transactions per minute, with additional throughput available as nodes are added to the cluster. Larger CPU configurations were not evaluated as part of this validation. This solution brief describes the validated architecture, use cases, performance test results, bill of materials, and deployment guidance for running Microsoft SQL Server 2025 on Lenovo ThinkAgile HX V4 and FX V4 hyperconverged systems.

SQL Server 2025 creates an opportunity to turn the enterprise data estate into an AI-ready platform with a native vector data type, AI model integration, and Azure Arc-enabled hybrid capabilities ( Microsoft SQL Server 2025 ). The refresh window is already open: SQL Server 2016 reaches the end of extended support on July 14, 2026, while SQL Server 2019 has been in extended support since February 28, 2025 (Microsoft Lifecycle: SQL Server 2016 ; Microsoft Lifecycle: SQL Server 2019 ). Lenovo provides a practical path forward through Database Validated Architectures on ThinkSystem V4 and a choice of ThinkAgile MX, VX, and HX platforms, helping organizations modernize from the ground to the cloud without forcing every workload into a single operating model.

1000's of engineers, architects, and platform teams will gather at the Venetian Convention and Expo Center in Las Vegas, Aug 31–Sept 3, 2026

The collaboration pairs Saudi-based data center infrastructure with Napster's enterprise AI platform, providing a fully in-Kingdom path to deploy agentic AI

System Infrastructure Software Market is expanding with rising hybrid cloud adoption, AI-powered IT operations, and enterprise investments in resilient digital infrastructure.

As artificial intelligence moves from experimentation to enterprisewide deployment, sovereignty concerns now touch every layer of the AI stack. Organizations are no longer evaluating AI based solely on model performance, innovation speed, or cost. They are increasingly asking whether they can control how AI systems are built, governed, operated, and evolved.

System Infrastructure Software Market is expanding with rising hybrid cloud adoption, AI-powered IT operations, and enterprise investments in resilient digital infrastructure.

Lenovo ThinkAgile MX650a V4, powered by NVIDIA RTX PRO 6000 Blackwell Server Edition GPUs, and Azure Arc-enabled Foundry Local combine to deliver a validated, on-premises platform for enterprise generative AI. The solution keeps models and prompts on customer-owned infrastructure while extending Microsoft Entra ID and Azure RBAC governance to the data center and the edge, letting regulated industries, edge sites, and cost-sensitive workloads run AI inference with predictable economics, low deterministic latency, and resilience through disconnected periods. Jointly engineered by Lenovo and Microsoft, the platform gives enterprises a proven path from AI pilot to production wherever their data resides.

With rising costs and complexity, it is becoming increasingly harder to manage IT infrastructure in a data center to address private cloud, hybrid cloud and container workloads. As it changes over…

World Hyperconverged Integrated System market forecast to 2035: CAGR in mid-to-high teens, driven by edge computing, AI workloads, and hybrid-cloud adoption. Key players include De

Building data centers isn't civil engineering anymore; it’s high-stakes product assembly where a disconnected software tool can create a $500M blind spot.

Enterprises evaluating large language model (LLM) inference infrastructure frequently assume they must invest in expensive, difficult to acquire, and costly to maintain GPUs and that CPU-only systems…

AI is shifting from responding to persisting. Discover why long-horizon reinforcement learning is the new frontier and the bottleneck of the decade.