Nvidia has unveiled its next generation GPU, the Rubin series, boasting marked performance improvements for AI inferencing.
Rubin is a new class of GPU, according to Nvidia, one that is purpose-built for massive-context processing, such as extensive software coding and high-speed video generation.
The new GPU will work “hand-in-hand” with the Vera CPU and be housed inside the latest Nvidia Vera Rubin NVL 144 CPX platform, the company said. This is an integrated NVIDIA MGX system that packs 8 exaflops of AI compute to provide 7.5x more AI performance than NVIDIA’s GB300 NVL72 systems.
It also promises 100TB of fast memory and 1.7 petabytes per second of memory bandwidth in a single rack.
However, the tech giant notes that Rubin CPX will be offered in other flexible configurations for customers who want to resume their existing infrastructure.
“The Vera Rubin platform will mark another leap in the frontier of AI computing — introducing both the next-generation Rubin GPU and a new category of processors called CPX,” said Jensen Huang, founder and CEO of Nvidia.
“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.”
What you need to know about the Rubin CPX
Rubin CPX is set for general release in 2026 and could further solidify Nvidia’s dominance in the GPU market. But its main aim appears to be to solve the challenge of deploying inference at scale.
It uses a ‘cost-efficient’ monolithic die design that’s packed with NVFP4 computing resources, the company said, which has been optimized to deliver extremely high performance and energy efficiency for AI inference tasks.
A key point Nvidia made during its presentation was that Rubin would enable the highest performance and token revenue for long context processing; the company said it was capable of going “far beyond what today’s systems were designed to handle”.
With Rubin, AI coding assistants will become sophisticated systems capable of comprehending and optimizing large-scale software projects, Nvidia claims.
American generative AI company Runway, which will use Nvidia hardware in its video offerings called Rubin a “major leap in performance”, highlighting it as a powerful tool for creators.
“Video generation is rapidly advancing toward longer context and more flexible, agent-driven creative workflows,” said Cristóbal Valenzuela, CEO of Runway.
“We see Rubin CPX as a major leap in performance, supporting these demanding workloads to build more general, intelligent creative tools. This means creators — from independent artists to major studios — can gain unprecedented speed, realism, and control in their work.”
Make sure to follow ITPro on Google News to keep tabs on all our latest news, analysis, and reviews.
MORE FROM ITPRO
- Jensen Huang says 'the AI race is on' as Nvidia shrugs off market bubble concerns
- How Nvidia took the world by storm
- Jensen Huang thinks the UK has immense AI potential
-
Nvidia braces for a $5.5 billion hit as tariffs reach the semiconductor industryNews The chipmaker says its H20 chips need a special license as its share price plummets
-
“The Grace Blackwell Superchip comes to millions of developers”: Nvidia's new 'Project Digits' mini PC is an AI developer's dream – but it'll set you back $3,000 a piece to get your hands on oneNews Nvidia unveiled the launch of a new mini PC, dubbed 'Project Digits', aimed specifically at AI developers during CES 2025.
-
Intel just won a 15-year legal battle against EUNews Ruled to have engaged in anti-competitive practices back in 2009, Intel has finally succeeded in overturning a record fine
-
Jensen Huang just issued a big update on Nvidia's Blackwell chip flawsNews Nvidia CEO Jensen Huang has confirmed that a design flaw that was impacting the expected yields from its Blackwell AI GPUs has been addressed.
-
How Nvidia took the world by stormAnalysis Riding the AI wave has turned Nvidia into a technology industry behemoth
-
Meta unveils two new GPU clusters used to train its Llama 3 AI model — and it plans to acquire an extra 350,000 Nvidia H100 GPUs by the end of 2024 to meet development goalsNews Meta is expanding its GPU infrastructure with the help of Nvidia in a bid to accelerate development of its Llama 3 large language model
-
TD Synnex buoyed by UK distribution deal with NvidiaNews New distribution agreement covers the full range of Nvidia enterprise software and accelerated computing products
-
PCI consortium implies Nvidia at fault for its melting cablesNews Nvidia said the issues were caused by user error but the PCI-SIG pointed to possible design flaws
