""Vera Rubin is designed to address this fundamental challenge that we have: The amount of computation necessary for AI is skyrocketing.""
""Today, I can tell you that Vera Rubin is in full production.""
""As you start to enable new types of workflows, like agentic AI or long-term tasks, that puts a lot of stress and requirements on your KV cache,""
Nvidia officially launched the Vera Rubin computing architecture at CES and confirmed it is already in production with expected ramp in the second half of the year. The architecture replaces Blackwell and follows Nvidia’s iterative hardware lineage through Hopper and Lovelace. Rubin comprises six coordinated chips centered on a Rubin GPU and includes improvements in storage (Bluefield), interconnection (NVLink), and a new Vera CPU for agentic reasoning. Major cloud providers and partners including Anthropic, OpenAI, and AWS plan Rubin deployments. Rubin systems will be used in HPE’s Blue Lion and the Doudna supercomputer at Lawrence Berkeley National Lab.
Read at TechCrunch
Unable to calculate read time
Collection
[
|
...
]