NVIDIA Unveils Vera Rubin Architecture to Scale AI Factories Worldwide
Summary:At the OCP Global Summit, NVIDIA unveiled the updated Vera Rubin AI factory architecture, designed to accelerate AI infrastructure deployment through open standards and modular design. The architecture introduces new liquid cooling technologies, modular I/O designs, and higher-density rack configurations, enhancing AI factory efficiency and scalability. Additionally, NVIDIA and partners announced open standards for next-generation AI factories, including AI factory design specifications and grid
NVIDIA Unveils Vera Rubin Architecture to Scale AI Factories Worldwide
At the OCP Global Summit, NVIDIA announced the updated Vera Rubin AI factory architecture, designed to accelerate AI infrastructure deployment through open standards and modular design. Here are the key highlights of the announcement:
1. Open Standards and AI Factory Design
NVIDIA, along with several partners, published open standards for next-generation AI factories, including AI factory design specifications and grid resilience standards. These standards provide a unified technical blueprint for global AI infrastructure deployment, covering aspects from power architecture to cooling infrastructure to physical layout.
2. Updates to the Vera Rubin AI Factory Architecture
- Liquid Cooling Technologies: The Vera Rubin architecture introduces rack-level blind-mate liquid cooling, interoperable rack-scale interconnect infrastructure, and a 5000A liquid-cooled busbar designed for high-power AI systems.
- Modular Design: The MGX 1.2 specification extends modularity with configurable I/O bays, fanless liquid-cooled compute trays, standardized power and cooling interfaces, and a common infrastructure that simplifies manufacturing and field service processes.
- Vera CPU Rack: Built on the third-generation NVIDIA MGX rack architecture, the Vera CPU rack provides a purpose-built platform for the agentic AI era, supporting tool calling, orchestration, and sandboxed execution to enable AI agents to interact with applications, data, and services in real time.
3. Partner Innovations
- Visionbay.ai: Introduced K-1, an AI factory platform based on an 800VDC architecture, combining NVIDIA's 800VDC reference architecture with power and thermal technologies from Vertiv to create a scalable foundation for high-density AI infrastructure.
- Meta and NVIDIA: Meta announced the adoption of NVIDIA Spectrum-6 102.4Tb/s Ethernet switches and the use of Spectrum-X Multiplane technology to achieve scalability and resilience in AI factory networks.
- Cisco and NVIDIA: Announced the USFP (Ultra-Dense Small Form Factor Pluggable) standard, a new optical pluggable transceiver module form factor designed to improve bandwidth density and power efficiency in AI factory networks.
Technical Mechanism Analysis
The core of the Vera Rubin architecture lies in its open standards and modular design. By introducing liquid cooling technologies and higher-density rack designs, NVIDIA has not only improved the energy efficiency of AI factories but also simplified deployment and maintenance processes. Furthermore, the MGX 1.2 specification's modular I/O design allows AI factories to scale flexibly according to demand without the need for large-scale infrastructure reconstruction.
Engineering Trade-offs and Actual Performance
Although the Vera Rubin architecture excels in energy efficiency and scalability, the implementation of liquid cooling technologies may require additional hardware investment and infrastructure transformation. Additionally, while the modular design enhances flexibility, it may also increase system complexity, necessitating more meticulous engineering management.
Developer Implementation and Deployment Recommendations
For developers planning to deploy AI factories, NVIDIA recommends adopting open standards to ensure compatibility with existing infrastructure and leveraging the modular design of the MGX 1.2 specification to simplify the deployment process. Additionally, developers can refer to Visionbay.ai's K-1 platform design, combining NVIDIA's liquid cooling technologies and power management solutions to achieve efficient and scalable AI infrastructure.
Conclusion
NVIDIA's Vera Rubin architecture, through open standards and modular design, provides new impetus for the global scale-up of AI infrastructure. Its innovations in energy efficiency, scalability, and flexibility will drive the widespread application of AI factories worldwide.
— END —Source: NVIDIA AI News (2026-10-12)
Tags: #NVIDIA #AI Infrastructure #Vera Rubin #Open Standards #Liquid Cooling
Editorial & Fact-Checking Note: This article is compiled from primary research, official release documentation, and source papers by the ZICQ Newsroom pipeline with automated entity verification and human editorial review. If you notice any technical inaccuracy, please submit a correction via our corrections policy or email our editorial desk directly.
Community Comments