Rubin is NVIDIA’s top-shelf Rubin-generation datacenter GPU. Each Rubin GPU will have:1
- 2 reticle-limited GPUs
- 50 PF FP4
- 288 GB HBM4
R200 will be released in 2H2026.1
From Nvidia Rubin revealed as Blackwell successor, powerful Vera CPU coming too:
Quote
set to splashdown in 2026, bringing support for 8-Hi HBM4 stacks
Quote
Other highlighted qualities of the Rubin platform include the speedy NVLink 6 Switch performance of up to 3,600 GB/s, and a touted CX9 SuperNIC component offering up to 1,600 GB/s.
From Nvidia’s Jensen Huang, Ian Buck, and Charlie Boyle on the future of data center rack density:
Quote
In the second half of 2026, Nvidia is promising to deliver the Vera Rubin NVL144, with a new Arm chip and a new GPU. The company has yet to disclose how much power it expects that rack to consume
Quote
From Rubin on, it will instead show the number of reticle-sized GPUs. Both Blackwell and Rubin are each made of two reticle-sized GPUs, which is why the number has doubled to 144, but in reality, the number of GPUs is the same over both generations.
VR200 NVL144
Starting with R200, the NVL nomenclature will change to reflect the number of reticle-limited GPU dies are in an NVL domain rather than the number of GPU packages. VR200 NVL144 will still have 72 GPU packages as GB200 NVL72, and each package will have two reticle-limited B200 GPUs.1
Each VR200 NVL144 will have:
- 3.3x GB300 NVL72
- 3.6 EF FP4 for inference
- 1.2 EF FP8 for training
- 13 TB/s HBM4
- 75 TB “fast memory”
- 250 TB/s NVLink6
- 28.8 TB/s ConnectX-9
VR200 NVL576
The VR200 NVL576 system will have1
- 576 Rubin GPUs, 15 EF FP4
- 2,304 Memory Chips (150 TB) at 4,600 PB/s
- 144 NVLink Switches at 1,500 PB/s
- 1,300 trillion transistors
- 12,672 Vera CPU cores
- 576 ConnectX-9 NICs
- 72 BlueField DPUs