NVIDIA Kicks Off the Next Generation of AI With Rubin — Six New Chips, One Incredible AI Supercomputer
View original at globenewswire.comNVIDIA Kicks Off the Next Generation of AI With Rubin — Six New Chips, One Incredible AI Supercomputer News Summary: The Rubin platform harnesses extreme codesign across hardware and software to deliver up to 10x reduction in inference token cost and 4x reduction in number of GPUs to train MoE models, compared with the…
Ce que nous avons tiré de cette source
The claims Via News extracted from this document. We point to the source; we don't replace it.
Spectrum-X Ethernet Photonics switch systems deliver 5x improved power efficiency and uptime
80% confidenceMicrosoft is building world's most powerful AI superfactories to serve any workload anywhere with maximum performance and efficiency, and with NVIDIA Vera Rubin GPUs will empower developers to create, reason and scale in new ways
80% confidenceThe NVIDIA Rubin platform helps keep scaling progress so advanced intelligence benefits everyone
80% confidenceNVIDIA Rubin platform represents major leap forward in AI infrastructure for handling massive token volumes and multistep reasoning
80% confidenceRubin platform delivers 4x reduction in number of GPUs needed to train MoE models compared to Blackwell platform
80% confidenceAWS and NVIDIA have been driving cloud AI innovation together for more than 15 years, and NVIDIA Rubin on AWS represents continued commitment to delivering cutting-edge AI infrastructure
80% confidenceIntelligence scales with compute, and when more compute is added, models get more capable, solve harder problems and make bigger impact for people
80% confidenceEfficiency gains in NVIDIA Rubin platform represent infrastructure progress that enables longer memory, better reasoning and more reliable outputs
80% confidenceLenovo is leveraging Neptune liquid cooling and global scale to help enterprises build AI factories that serve as intelligent, accelerated engines for innovation
80% confidenceNVIDIA's Rubin platform promises to deliver step-change in performance and efficiency required to deploy most advanced models to billions of people
80% confidenceRubin arrives at exactly the right moment, as AI computing demand for both training and inference is going through the roof
80% confidenceWith annual cadence of delivering new generation of AI supercomputers and extreme codesign across six new chips, Rubin takes a giant leap toward the next frontier of AI
80% confidenceNVIDIA Rubin will be a rocket engine for AI and is the gold standard infrastructure for training and deploying frontier models at scale
80% confidenceSpectrum-X Ethernet Photonics delivers 10x greater reliability for AI applications
80% confidenceRAS Engine modular cable-free tray design enables up to 18x faster assembly and servicing than Blackwell
80% confidenceNVIDIA Rubin platform represents important advancement for reasoning, agentic and large-scale inference workloads
80% confidenceWith NVIDIA Vera Rubin platform, HPE is building next generation of secure, AI-native infrastructure to turn data into intelligence and enable enterprises to become AI factories
80% confidenceWith gigascale AI factories powered by NVIDIA Vera Rubin architecture, OCI gives customers infrastructure foundation to push limits of model training, inference and real-world AI impact
80% confidenceRubin platform delivers up to 10x reduction in inference token cost compared to Blackwell platform
80% confidenceGoogle is proud of deep and long-standing relationship with NVIDIA and collaboration will continue bringing impressive capabilities of Rubin platform to customers
80% confidence