NVIDIA has unveiled its latest enterprise offering, RTX PRO Servers, aimed at speeding the shift in IT infrastructure towards AI factories. Announced at COMPUTEX, this launch marks a significant step in transitioning traditional CPU-based enterprise systems to full-stack, GPU-accelerated platforms powered by the NVIDIA Blackwell architecture.
At the heart of this development is the new RTX PRO 6000 Blackwell Server Edition GPU, which promises high performance and energy efficiency. The servers are designed to support a wide range of enterprise applications, including AI workloads, design and engineering simulations, and business operations. In tandem, NVIDIA introduced its Enterprise AI Factory validated design—a framework that combines servers, networking, storage, and software for enterprises looking to implement AI-native infrastructure on premises.
Global partners adopt NVIDIA’s enterprise AI factory model
Major industry players including Cadence, Foxconn and Lilly have already committed to building their own AI factories using NVIDIA’s new design and RTX PRO Servers. These factories are set to enhance product development, simulate engineering processes, and run AI-driven business operations.
Foxconn, in particular, is adopting the technology to accelerate its transformation across manufacturing, electric vehicles, healthcare, and robotics. “Foxconn is harnessing the performance of NVIDIA Blackwell to build AI infrastructure that will transform every facet of electronics manufacturing,” said Young Liu, chairman of Foxconn. “Through our close collaboration with NVIDIA, we will accelerate the integration of AI across our global operations and deliver smarter electronics for the world.”
According to NVIDIA CEO Jensen Huang, “AI is revolutionising every industry — every company will build or rent AI factories to run their businesses and power the intelligence of their products. With our global partner ecosystem, we’re helping enterprises infuse AI into their workforce, automate their factories and build AI-native products.”
Full-stack infrastructure built for accelerated enterprise workloads
The RTX PRO 6000 Blackwell GPU serves as a universal data centre platform capable of supporting diverse and complex workloads. Each RTX PRO Server can be configured with up to eight GPUs and integrated with NVIDIA BlueField-3 DPUs and ConnectX-8 SuperNICs for optimal performance and connectivity.
Through the NVIDIA AI Enterprise software suite, companies can scale and refine their AI initiatives using tools such as NIM and NeMo microservices. The servers are also optimised for creating digital humans, running AI search tools, and executing simulations through the NVIDIA Omniverse platform.
The Enterprise AI Factory validated design helps enterprises deploy AI systems faster with recommended configurations from NVIDIA Enterprise Reference Architectures. These setups include NVIDIA-Certified Servers, Spectrum-X networking, BlueField DPUs, certified storage options, and NVIDIA AI Enterprise software, creating a consistent and scalable AI foundation.
A growing ecosystem supporting AI factory deployments
NVIDIA’s announcement also includes support from a wide ecosystem of global partners. Major system providers like Cisco, Dell Technologies, Hewlett Packard Enterprise, and Lenovo are set to offer full-stack solutions featuring RTX PRO Servers. Additional partners, including ASUS, MSI, Supermicro, Quanta Cloud Technology, and Foxconn, are developing compatible server systems to meet enterprise needs.
Storage partners such as Dell, IBM, Pure Storage, and NetApp are also contributing certified solutions to ensure smooth integration and scalability. Consulting firms including Accenture, Deloitte, EY, Infosys, Tata Consultancy Services, and Wipro are working with businesses to adopt the Enterprise AI Factory framework using Blackwell-accelerated data centres.
The infrastructure supports a broad range of software platforms such as Ansys, CrowdStrike, Elastic, Red Hat, Siemens, and Synopsys, making it adaptable to different business environments and operational demands.
