NVIDIA Blackwell

NVIDIA Blackwell is a GPU architecture from NVIDIA for data centers and AI that succeeded the Hopper architecture.

2 articles
Last mentioned

NVIDIA Blackwell is a GPU architecture that NVIDIA announced in March 2024 as the successor to its Hopper generation. It is named after the American mathematician and statistician David Blackwell. The architecture underpins data center products such as the B200 and GB200 as well as consumer graphics cards in the GeForce RTX 50 series. The GB200 is a superchip that pairs a Grace CPU with Blackwell GPUs, and the GB300 uses the higher-performance Blackwell Ultra.

It was designed primarily for training and running large AI models, and cloud providers rent out GPUs based on it.

This entry is based on AIPOST articles and widely known facts. If something is wrong, please send us a correction request.

Articles covering this entry

…atible GPUs can reduce assignment delays, but testing found inference on a Blackwell MIG slice—a portion of a GPU allocated as a separate instance—was 1.8 times…

Elon Musk said xAI's Colossus 2 facility in Memphis has deployed 550,000 NVIDIA GB200 and GB300 graphics processing units, or GPUs, the specialized chips used f…


© 2026 AIPOST. All rights reserved.

AIPOST is an AI publication covering practical AI, AI security, performance, startups, health, ethics and industry news. No account is needed, and our privacy policy explains how we handle personal information.