NVIDIA Unveils Next-Gen GPU: Inference Computing Power Soars to Five Times That of Blackwell
1 week ago / Read about 0 minute
Author:小编   

On January 5 (local time), Jensen Huang, the CEO of NVIDIA, took the stage at CES to announce the launch of the company's next-generation AI platform, Rubin. This cutting-edge platform encompasses six brand-new chips: the Vera CPU, Rubin GPU, NVLink 6 Switch, ConnectX-9 SuperNIC, BlueField-4 DPU, and Spectrum-6 Ethernet Switch.

Among these, the standout is the Rubin GPU chip, which is outfitted with a third-generation Transformer engine. This advanced engine enables the chip to achieve an impressive NVFP4 inference computing power of up to 50 PFLOPS, marking a fivefold increase over the capabilities of Blackwell.

At present, NVIDIA's Rubin platform has transitioned into full-scale production. Products built on this innovative platform are projected to be accessible through partners starting in the second half of 2026.