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.

Nvidia will also offer a single rack called the GB300 NVL72 that offers 1.1 exaflops of FP4, 20TB of HBM memory, 40TB of “fast memory,” 130TB/sec of NVLink bandwidth and 14.4 TB/sec networking.

Late into his keynote at Nvidia’s 2025 GTC conference, CEO Jensen Huang announced the next generation of Nvidia AI data center chips named Vera and Rubin, after the groundbreaking astrophysicist. Vera is an Arm-based CPU chip that is the ringleader in an Nvidia AI server,