Red Hat has announced a major expansion of its collaboration with NVIDIA to accelerate enterprise AI innovation by integrating enterprise open source software with rack-scale AI infrastructure. The partnership centers on Red Hat Enterprise Linux for NVIDIA, a new specialized edition of the company’s flagship operating system, designed to deliver Day 0 support for NVIDIA’s latest Rubin platform. This initiative aims to provide a stable, secure, and high-performance foundation for production-grade AI deployments across hybrid cloud environments. The NVIDIA Vera Rubin platform, featuring the new Vera CPU, BlueField-4 data processor, and the NVL72 rack-scale solution, is engineered to power the next generation of agentic AI and advanced reasoning systems. Red Hat is aligning its entire AI and hybrid cloud portfolio—including Red Hat OpenShift and Red Hat AI—to support this platform from launch, ensuring enterprises can scale AI workloads with confidence. Red Hat Enterprise Linux for NVIDIA will be optimized for the Rubin platform, offering immediate support for the latest hardware innovations. The new edition maintains full compatibility with the mainline Red Hat Enterprise Linux build, allowing customers to transition seamlessly to the standard version as their production needs evolve, without sacrificing performance or application compatibility. Key capabilities of the partnership include enhanced security through NVIDIA Confidential Computing, which provides cryptographic proof that sensitive AI workloads—such as model data and memory—are protected across GPUs and system memory. Red Hat OpenShift will integrate NVIDIA’s infrastructure software and CUDA X libraries, enabling automated deployment, configuration, and lifecycle management of accelerated workloads. Support for the BlueField-4 DPU will also improve networking, cluster management, and resource efficiency, delivering a consistent, enterprise-grade experience. Red Hat AI will expand its integrations with NVIDIA, adding support for distributed inference using open-source models from NVIDIA, including those focused on vision, robotics, and industry-specific applications. The platform will also support the Red Hat AI Inference Server, Red Hat Enterprise Linux AI, and Red Hat OpenShift AI, creating a unified, production-ready environment for AI development and deployment. The new Red Hat Enterprise Linux for NVIDIA will be available in the second half of 2026, coinciding with the general availability of the NVIDIA Vera Rubin platform. Customers will be able to access the latest drivers, tools, and support through the Red Hat Customer Portal. This collaboration marks a significant step in bringing industrial-strength open source to the forefront of AI infrastructure. As organizations shift from AI experimentation to full-scale production, the need for reliable, secure, and scalable platforms has never been greater. By combining Red Hat’s enterprise-grade open source foundation with NVIDIA’s cutting-edge hardware, the two companies are setting a new standard for how enterprises can build, deploy, and manage AI at scale.