US EditionNvidia Offers First Look at Vera Rubin Superchip
The Vera Rubin superchip will power two Argonne supercomputers delivering 2,200 exaflops to accelerate scientific and national security AI breakthroughs.
- At GTC, NVIDIA unveiled the Vera Rubin superchip, with Jensen Huang saying `This is the next generation Rubin`; NVIDIA is preparing Rubin for production next year.
- At Argonne National Laboratory, NVIDIA, Oracle and the U.S. Department of Energy will build the largest AI supercomputer featuring 100,000 Blackwell GPUs to enable agentic AI workflows using the NVIDIA Megatron-Core library and NVIDIA TensorRT inference stack.
- A single-board system combines an 88-core Vera CPU with twin Rubin GPUs, each with 288GB HBM4 memory and linked via NVLINK-C2C; NVL144 delivers 3.6 Exaflops FP4 inference and 1.2 Exaflops FP8, 3.3 times faster than GB300 NVL72.
- At the beginning of October, Microsoft delivered the GB300 NVL72 cluster for OpenAI, while industry demand could reach about 12 million Blackwell + Rubin chips and Nvidia's CFO clarified $500 billion datacenter revenue with Morgan Stanley modeling ~$407B implying a $93B upside.
- Equinox will include 10,000 Blackwell GPUs and be available in the first half of 2026, with both systems interconnected to deliver 2,200 exaflops; Nvidia teased Rubin Ultra NVL576 and a super-sized Rubin for the first of 2027.
12 Articles
12 Articles
News from HD Technology. Visit www.hd-tecnologia.com for the latest news. At GTC 2025, Jensen Huang took the stage to show for the first time the Superchip Vera Rubin, a platform that promises to mark the start of the next generation of hardware for AI. This new architecture combines a Vera CPU with two massive Rubin GPUs, both designed to deliver extreme performance in training tasks and inference of advanced models. A look at the most powerful…
NVIDIA's "Vera Rubin" Superchip System Pictured for the First Time
NVIDIA yesterday had a GTC conference in Washington D.C., showing its latest "Vera Rubin" Superchip symbiote. Pictured for the first time is the combination of two "Rubin" GPU paired with a single "Vera" CPU carrying 88 custom NVIDIA cores and 176 threads in a single package. NVIDIA quoted performance targets of roughly 50 PetaFLOPS of FP4 compute per Rubin GPU, which yields about 100 PetaFLOPS FP4 for the two-GPU Superchip. The company said eng…
Here's NVIDIA’s Vera Rubin AI Superchip — 88 Cores, Two GPUs, Gobs Of Memory And Next-Level Design
NVIDIA held its annual Graphics Technology Conference (GTC) in Washington D.C. yesterday, and as a surprise showing in the middle of his keynote, company CEO Jensen Huang pulled out a Vera Rubin Superchip, marking the first time that this product was shown to the public. The part looks quite different from even the GB300 Blackwell Ultra Superchip,
technewstube.com
Tom's HardwareNvidia reveals Vera Rubin Superchip for the first time — incredibly compact board features 88-core Vera CPU, two Rubin GPUs, and 8 SOCAMM modules
First images of Nvidia's Vera Rubin Superchip depicting two Rubin GPUs and a multi-chiplet Vera CPU surrounded by SOCAMM2 memory modules emerge.
Coverage Details
Bias Distribution
- 100% of the sources lean Left
Factuality
To view factuality data please Upgrade to Premium
