bit.fanBit.FanWeb3 News · Betting Reviews · Market Insights
‹ BackNewsVera Rubin

Vera Rubin

Nvidia
2026-07-22 04:18:11

Nvidia Rubin testing shifts AI hardware trade back toward supply concerns

A sharp sell-off in U.S. AI stocks had recently been tied to concerns that Kimi K3, a high-performance and lower-cost open-weight model, could weaken confidence in the expensive closed-model path pursued by major U.S. AI companies. That view hit semiconductor names last week, pushing the PHLX Semiconductor Index briefly into technical bear-market territory and dragging Nvidia, Micron, equipment makers, and memory stocks lower. But a new report from The Information gave investors a different set of numbers to focus on. Nvidia’s next-generation Vera Rubin server systems have entered customer testing, with CoreWeave, Microsoft, OpenAI, Anthropic, and SpaceXAI among dozens of customers that have received a small number of test racks. Each Rubin rack includes 72 GPUs and is priced at roughly $7 million to $8 million, above the roughly $5 million price tag for the current Grace Blackwell 300 rack. Nvidia senior vice president of hardware engineering Andrew Bell said more than 10 manufacturing partners are ultimately expected to gain the capacity to produce as many as 1,000 Rubin racks per day. That theoretical output, multiplied by rack pricing, points to a massive infrastructure buildout still underway, even as the market debates whether lower-cost AI models will reduce chip demand or expand deployment enough to keep spending elevated.

1340
Nvidia Rubin testing shifts AI hardware trade back toward supply concerns
NVIDIA
2026-07-21 15:48:04

NVIDIA unveils Spectrum-6 Ethernet switching system for next-generation hyperscale AI factories

NVIDIA on July 21 introduced its new Spectrum-6 Ethernet switching system, designed to support hyperscale AI infrastructure built on the Vera Rubin platform. The company said the system delivers 102.4 Tbps of switching capacity per device, twice that of the previous generation. It is intended to connect hundreds of thousands of GPUs and CPUs and provide networking support for large-scale AI training, agent-based applications, and inference workloads. NVIDIA also said Spectrum-6 will serve as a core component of the next-generation Spectrum-X Ethernet platform. According to the company, the product is built to improve overall computing efficiency in AI factories through high bandwidth, intelligent traffic management, and fault recovery capabilities. CoreWeave, Microsoft, Nebius, SpaceXAI, and Tesla are set to be among the first infrastructure builders to deploy the system.

1870
NVIDIA unveils Spectrum-6 Ethernet switching system for next-generation hyperscale AI factories
Nvidia
2026-07-21 15:01:28

Nvidia says key customers have started testing Vera Rubin systems

Nvidia said its major customers have begun testing Vera Rubin equipment, according to a Bloomberg report cited by BlockBeats on July 21. The company also said its new Vera processor is faster than AMD's Turin. Nvidia added that the chips are being delivered on schedule and are intended for use in artificial intelligence data centers. The update points to progress in the rollout of Nvidia's next-generation hardware for AI infrastructure, based on the company's own statements and the Bloomberg report referenced in the original brief.

1800
Nvidia says key customers have started testing Vera Rubin systems
South Korea
2026-07-19 07:09:40

South Korea plans roughly KRW 4 trillion push to buy 10,000 Nvidia Vera Rubin GPUs next year

South Korea plans to spend about KRW 4 trillion next year to purchase 10,000 of Nvidia’s latest Vera Rubin GPUs, according to Odaily. The chips are intended to support the expansion of public-facing services such as a nationwide AI initiative and to help build out national AI infrastructure. The Ministry of Science and ICT plans to extend its project for strengthening the foundation for AI computing resource utilization, with all GPUs introduced next year set to use the Vera Rubin platform. Under the program, private-sector companies including cloud service providers that run data centers would handle the full process covering GPU procurement, construction, and service operations, with funding coming from the government budget. Over the past two years, the South Korean government has secured more than 30,000 GPUs through this project and related measures, supplying them to both public and private sectors. The input states that compared with the previous-generation Blackwell architecture, Vera Rubin can deliver up to six times higher training performance and more than eight times better inference performance. A ministry official said the GPU acquisition plan and budget size have not been finalized, while industry sources pointed to limited spare data center capacity and said securing large-scale new resources before next year would not be easy.

1830
South Korea plans roughly KRW 4 trillion push to buy 10,000 Nvidia Vera Rubin GPUs next year