vCluster Labs ships Stacks to package any AI environment as a managed service. Oct 8, 2026 SAN FRANCISCO - October 8, 2026 - vCluster Labs, the platform operators use to build and run their own AI cloud like a hyperscaler, today announced the general availability of Stacks in vCluster Platform v4.12, which automate the long list of ordered setup steps required before a managed service can go live, so an operator can deploy the same environment to every tenant. The limit on how many managed services an operator can offer is not GPU supply. It is the setup work between an empty tenant cluster and an environment that is ready to use, stitched together by hand and repeated for every tenant. "Every operator runs into the same wall," said Lukas Gentele, Co-Founder and CEO of vCluster Labs. "Once the cluster exists, the real project starts: registry credentials, ingress, certificates, GPU components, a control plane, and registration, all wired together by hand, in the right order, again for every tenant." A Stack is that environment described once: which applications, in what order, what has to be healthy before the next thing starts, and which parameters the person deploying it is allowed to set. Deploy it and a tenant gets a running environment. Deploy it a hundred times and the hundredth costs what the first one did. A new managed service becomes something an operator ships rather than something they staff. Stacks make new services faster to launch and easier to operate. A new managed service can go live in days instead of months, because the work is authoring one template rather than staffing an install for every tenant. Each additional tenant is a deployment rather than a project, which is how a GPU fleet turns into higher-margin managed products instead of raw capacity sold by the hour. Operating them is the other half. Health, logs, upgrades, and clean tenant removal are handled by the platform, and because every tenant runs an identical environment, the team supporting the fiftieth tenant is troubleshooting the same configuration it learned on the first. Operators add new tenants without adding people to keep each environment running. Get started with the #1 platform for AI infra. Trusted by today's fastest-growing AI cloud builders. Available today. Stacks ship with ready-made examples covering the environments operators are standing up now: * NVIDIA Run:ai, in four certified configurations covering both a dedicated control plane for a single cluster and a shared control plane serving many tenant clusters. * NVIDIA Dynamo, standing up a distributed inference environment inside a tenant cluster so an operator can offer serving for large models that span many GPUs and nodes. * Saturn Cloud, installing the AI token factory platform inside a tenant cluster so an operator can offer per-token inference, fine-tuning, and usage billing on GPUs they already own. "Each of these is a service an operator can put in front of customers without building the integration first," Gentele added. "That is the difference between selling capacity and selling products, and you add one without adding a team to run it. The repository is open, so the next one does not have to come from us." "Adding a service to an AI cloud shouldn't mean starting a new integration project each time," said Hugo Shi, co-founder and CTO of Saturn Cloud. "vCluster built the Saturn Cloud Stack, so an operator can deploy it like any other service. Each customer gets private inference in an environment of their own, ready to use right away, and a small team can keep adding services without adding headcount." Anyone can author a Stack. Stacks are not limited to integrations vCluster Labs builds. Any team whose software runs inside customer clusters can describe a correct installation once and have it land identically everywhere it is deployed. A public repository is open for exactly that, so a team that has already solved this can hand the next team a starting point. Stacks are generally available today in vCluster Platform 4.12 with vCluster 0.37. Supporting resources. About vCluster Labs vCluster Labs is the platform operators use to build and run their own AI cloud, like a hyperscaler. vCluster turns raw GPUs into cluster products they can sell, from bare metal provisioning up through managed Kubernetes, Slurm and inference clusters, with tenant isolation at the infrastructure layer. The result is more margin per GPU, less dependence on a few big customers, and new managed services live in days instead of months. vCluster is trusted by fast-growing AI cloud providers including Nebius, Groq, Firmus and Corvex across 100K+ GPUs, and by enterprises including Adobe, Samsung and Deloitte. Table of Contents

The AI infrastructure bottleneck: why 'good enough' Kubernetes isn't cutting it anymore. While security eyes are on the RSAC conference in San Francisco this week, the compute world is focused on KubeCon EU in Amsterdam. But the theme of artificial intelligence is the pervasive across both, as in enterprise information technology we've reached a point where "AI curiosity" has officially been replaced by "AI urgency." Every chief information officer I talk to is under immense pressure to move from those neat little research-and-development experiments to actual production-grade deployment. But as they scale, they're hitting a wall that isn't about the models or the data - it's about the plumbing. Specifically, it's the graphics processing unit infrastructure bottleneck. For years, we've treated Kubernetes as the panacea to infrastructure woes. Need to scale? Throw it in a container. Need to orchestrate? K8s is your friend. But when you're dealing with Nvidia Corp. Blackwell B300s and massive training clusters, the standard way of doing things is sharing overprovisioned environments or waiting weeks for dedicated hardware. These are recipes for project failure, only adding to the narrative that the majority of AI project fail. Today at KubeCon, neocloud provider QumulusAI and vCluster, creators of virtual Kubernetes cluster technology, announced a partnership to address much of the friction between infrastructure agility and the rigid demands of high-performance GPUs. The real cost of infrastructure friction. Today's reality is that enterprise development teams are currently stuck in a "pick your poison" scenario. * The wait-and-see approach: A dedicated GPU environment is requested, but the IT team needs time to provision and tells the requester to check back in three weeks. In the past, this has been an annoyance but in the AI race, three weeks is an eternity and could be the difference in being an industry leader or a laggard. * The Wild West approach: Business units share a massively overprovisioned environment. It's faster to get into, but it's a security nightmare, and resource contention makes training runs highly unpredictable and ever harder to forecast when attempting to capacity plan. This inefficiency is more than just an inconvenience; it's a massive drain on return on investment, since time is money. When companies deal with hyperscalers or neocloud providers, they expect the kind of speed that Nvidia Blackwell B300s and RTXPRO 6000s promise. Having those chips sit idle while a developer fumbles a namespace configuration is the compute version of malpractice. QumulusAI and vCluster: partitioning power. The partnership between QumulusAI and vCluster brings customers a way to "slice and dice" high-end GPU power without the overhead of traditional virtualization. This gives customers more options but more importantly, the exact amount of GPU power they need to run their accelerated computing workloads, the primary one being AI. QumulusAI came to market with a value proposition of building a turnkey, vertically integrated AI cloud. Think of QumulusAI as a company that didn't just build a fast car, but designed the engine, the fuel and the highway it runs on. This "hyperspeed compute" setup provides massive power, but QumulusAI also provides the dashboard to keep all the horsepower under control. In fact, the company will let customers only use a piece of the engine if that's all that's required for the journey. By integrating vCluster's virtual Kubernetes technology, QumulusAI is essentially giving enterprises faster and more granular control of isolated environments. Instead of spinning up an entire physical cluster for every project, which is slow and expensive, teams can now spin up isolated virtual clusters on shared GPU hardware. This gives developers the "feel" of a dedicated environment - complete with their own application programming interface server and full control - while the platform team gets to maximize the utilization of those incredibly expensive GPUs. The vCluster AI Lab: innovation at the edge. Perhaps the most interesting part of this news is the launch of the vCluster AI Lab. The lab should provide QumulusAI customers assurance they can continue to use the platform for the long term. As the physical chips that are used for AI, such as GPUs, rapidly improve, the software managing them must stay ahead of the curve. This lab ensures that no matter how advanced the hardware becomes, the systems can handle the workload. It allows vCluster engineers to prototype how Kubernetes should handle emerging AI workloads in real time. Accelerating the move to AI factories. As I've noted in my previous posts, in 2026 the goal for companies should be to move AI factories from being projects to production infrastructure. To get there, organizations need three things: * Access: Getting the latest silicon (such as the B300) without a two-year lead time. * Isolation: Ensuring that Team A's training run doesn't crash Team B's inference model. * Speed: Moving from idea to environment in minutes, not months. This partnership addresses all three points and allows a midsized enterprise to act like a large company and enterprises to act like hyperscalers. They get the security of an isolated environment and the performance of bare-metal GPUs, all managed through a unified Kubernetes stack. Final thoughts. The AI race is going to be won by the companies that solve the operational headaches of GPU management. The technology is there, but can organizations deploy it in a way where it meets their needs now, doesn't break the bank and can scale with them? The partnership between QumulusAI and vCluster lowers the barrier to entry for secure, high-performance environments and makes it possible for AI teams to move as fast as their ideas. And in today's market, speed isn't just an advantage - it's the only thing that matters. Zeus Kerravala is a principal analyst at ZK Research, a division of Kerravala Consulting. He wrote this article for SiliconANGLE. Image: QumulusAI. A message from John Furrier, co-founder of SiliconANGLE: Support our mission to keep content open and free by engaging with theCUBE community. Join theCUBE's Alumni Trust Network, where technology leaders connect, share intelligence and create opportunities. * 15M+ viewers of theCUBE videos, powering conversations across AI, cloud, cybersecurity and more * 11.4k+ theCUBE alumni - Connect with more than 11,400 tech and business leaders shaping the future through a unique trusted-based network. About SiliconANGLE Media SiliconANGLE Media is a recognized leader in digital media innovation, uniting breakthrough technology, strategic insights and real-time audience engagement. As the parent company of SiliconANGLE, theCUBE Network, theCUBE Research, CUBE365, theCUBE AI and theCUBE SuperStudios - with flagship locations in Silicon Valley and the New York Stock Exchange - SiliconANGLE Media operates at the intersection of media, technology and AI. Founded by tech visionaries John Furrier and Dave Vellante, SiliconANGLE Media has built a dynamic ecosystem of industry-leading digital media brands that reach 15+ million elite tech professionals. Our new proprietary theCUBE AI Video Cloud is breaking ground in audience interaction, leveraging theCUBEai.com neural network to help technology companies make data-driven decisions and stay at the forefront of industry conversations.

vCluster Labs has introduced vMetal, a bare metal machine management layer designed to help neocloud providers and AI factories provision and operate GPU infrastructure at scale. The platform automates the lifecycle of bare metal GPU servers, from provisioning and assignment to upgrades and repurposing. Combined with vCluster's tenant orchestration and AI-focused Certified Stacks, the vCluster Platform delivers what the company calls the industry's first unified AI infrastructure stack. The platform spans physical machines, Kubernetes environments and AI-ready application stacks, deployable across public cloud, private data centres and GPU neocloud providers. The initial Certified Stacks feature deep integration with NVIDIA Run:ai, which has validated vCluster as Run:ai conformant. The new release is available now.