Building the hybrid AI factory of the future: Red Hat achieves AI Cloud Ready status for the NVIDIA Cloud Partner (NCP) program

Navigating the complexities of AI infrastructure shouldn’t be a barrier to innovation. Red Hat has completed the first phase of AI Cloud Ready status for the NVIDIA Cloud Partner (NCP) program to help address the increasing complexities of AI. NCPs build and operate GPU accelerated AI platforms to deliver and support full-stack, AI-optimized offerings based on the NCP software reference guide. This reference architecture is a proven blueprint for the full stack, including GPU servers, networking, storage, and software, to enable NCPs to deliver AI capacity as reliable, consistent services instead of custom one-off builds. 

By aligning with the NCP software reference architecture and validation specifications for ISVs, Red Hat provides a consistent foundation for repeatable, enterprise-scale AI deployments. This AI Cloud Ready status helps ensure that Red Hat technologies, such as Red Hat AI, interoperate predictably with the NCP blueprint as deployments scale. Organizations can more easily scale distributed inference, simplify model-to-data integration, and move from pilot to production across hybrid environments with reliable day-to-day operations. 

This transition to production-ready AI isn’t just a trend—it’s a decisive shift from experimental lab projects to core enterprise infrastructure. Red Hat’s collaboration with NVIDIA provides the co-engineered software foundation for this industrialization: transforming AI from a standalone experiment into an integrated, operationalized engine for business transformation.

Enabling Anything-as-a-Service (XaaS) frameworks

As a validated ISV, Red Hat enables NCPs to deliver cloud-native resources on-demand within their private datacenters. This move toward an “as-a-service” framework provides the high performance required for training and heavy inference workloads while replacing manual, one-off provisioning with standardized, industrial processes. Built on a security-focused foundation, Red Hat AI helps ensure these “as-a-service” patterns remain consistent to deploy and predictable to operate across hybrid environments.

Through the NCP program, Red Hat facilitates 5 critical service models for NCP partners:

• Bare Metal-as-a-Service: Automated provisioning of physical hardware for maximum performance and isolation.
• Cluster-as-a-Service: Rapid deployment of orchestrated Kubernetes environments via Red Hat OpenShift.
• VM-as-a-Service: Virtualized instances for workloads requiring specific isolation or legacy integration.
• GPU-as-a-Service: Intelligent orchestration and pooled access to expensive acceleration hardware to maximize use across bare metal or virtual machines (VMs).
• Model-as-a-Service: Centrally managed models delivered through standard APIs with multitenant governance. This enables controlled use, simpler application integration with enterprise data, and distributed inference to scale throughput and reduce latency across multiple nodes.

This framework gives organizations the flexibility of a public cloud without sacrificing security and sovereignty over data location, access controls, and operational governance. By hosting model services on Red Hat infrastructure, NCPs can capture efficiency gains directly rather than yielding that margin to a cloud provider. Additionally, organizations benefit from a vast ecosystem of hardware, software, and service partners that provide a unified, enterprise-grade foundation for AI across the hybrid cloud. This ecosystem allows organizations to move from experimentation to production with consistent operational rigor. This autonomy helps ensure NCPs maintain control over change windows, deployment cadences, and data privacy while preparing for the shift toward rack-scale architectures.  

The path to AI independence and sovereignty

The collaboration between Red Hat and NVIDIA offers the definitive path to AI independence and sovereignty. Red Hat delivers full-stack control over the deployment lifecycle, supporting privacy for prompts and sensitive data. This provides platform consistency across the hybrid cloud and enables token economics that can increase efficiency on the enterprise balance sheet. As adoption expands to agentic AI, Red Hat AI introduces an AgentOps approach to help move from experimentation to production operations with stronger observability and control.

This validation, as part of the NCP program, builds on our recent introduction of Red Hat AI Factory with NVIDIA and Red Hat and NVIDIA’s ongoing joint collaborative work to enable Day 0 support for NVIDIA hardware across the Red Hat portfolio. These efforts extend hybrid cloud capabilities to the next generation of AI factory-scale accelerated computing infrastructure. 

As the industry shifts from labs to enterprise-grade production, Red Hat, in collaboration with NVIDIA, provides the performance, scale, and stability for the next decade of innovation. We encourage IT leadership and platform engineers to review the latest technical datasheets and initiate proofs of concept to start their AI transformation today.

Ready to modernize your infrastructure? Explore these resources to begin your transition to industrialized AI:

• Learn more about Red Hat’s collaboration with NVIDIA.
• Get started today with the Red Hat AI Factory with NVIDIA.