NVIDIA Unveils Rubin CPX GPU for Massive-Context AI Models
The new processor class is purpose-built for million-token inference, delivering up to 30 petaflops of compute to accelerate generative video and coding.
A New Class of Processor
NVIDIA on September 9, 2025, introduced the Rubin CPX, a new class of graphics processing unit designed specifically for massive-context artificial intelligence. The chip is engineered to accelerate the inference process for models that handle million-token contexts, such as those used in advanced software coding and generative video applications. The announcement was made at the AI Infra Summit.
"Just as RTX revolutionized graphics and physical AI, Rubin CPX is the first CUDA GPU purpose-built for massive-context AI, where models reason across millions of tokens of knowledge at once,” said Jensen Huang, founder and CEO …
Archive Access
This article is older than 24 hours. Create a free account to access our 7-day archive.