Nvidia recently revealed details about its new AI graphics processors, and one of them is named for a scientist familiar to many space fans. Vera Rubin, the namesake for one of Nvidia's new AI GPU platforms,

On the near horizon — H2 2025 — Nvidia will release Blackwell Ultra, a GPU that’ll come in several configurations. A single Ultra chip will offer the same 20 petaflops of AI performance as Blackwell, but with 288GB of memory — up from 192GB in vanilla Blackwell.

The Vera Rubin die will have two GPUs sandwiched together, delivering over 50 petaflops of FP4 performance per chip. Nvidia claims that a full rack of these will deliver over 3.6 exaflops of performance—that's more than three times what's possible on existing Blackwell hardware.

Nvidia Corp., looking to cement its place at the heart of the artificial intelligence boom, laid out plans for more powerful chips, a model for robotics, and “personal AI supercomputers” that will let developers work on desktop machines.

Nvidia's DGX machines are designed for running large language models and other "AI" applications on the desktop. Shown off at its keynote were a Mac Mini-size device called the Spark, a motherboard, and a full-size desktop called the Station.

Nvidia chief Jensen Huang on Tuesday showcased cutting-edge chips for artificial intelligence and new applications for the technology, shrugging off talk of China's DeepSeek disrupting the market and