Premal Savla

[Jensen Huang @JensenHuang]

DGX-1 turns 10!

I delivered the first DGX-1 to OpenAI in SF on Aug 9, 2016.

On it, I wrote: "To the future of computing and humanity." It was a new kind of computer for a new kind of software.

Ten years later, I gave Elon a DGX Spark while visiting SpaceX. "From a single spark, a world of intelligence."
Spark is 1 PFLOP of NVFP4 at 200W — 5x DGX-1, in a tiny desktop computer.

Just imagine the next 10 years.

[NVIDIA AI Infrastructure @NVIDIAAIInfra]

ICYMI: An #NVIDIADGX AI supercomputer is now online at @NPS_Monterey.

The DGX GB300 with NVIDIA Mission Control software gives NPS students and faculty access to AI computing, including model training and inference capability for applications spanning weather prediction, cybersecurity, and response planning.

Thank you to our partners @DDNintelligence, @VAST_Data, and @Vertiv for helping to bring up this system with us.

Learn how we're helping to fuel AI innovation with NPS ➡️ https://t.co/Irph7JHY34

[Premal Savla @PremalSavla]

It's been a week of high honors for NVIDIA's leadership.

Founder and CEO Jensen Huang and Chief Scientist Bill Dally were recognized for their work in accelerated computing — the foundation of modern AI. https://t.co/8qFZUEWl5T

[Premal Savla @PremalSavla]

Explore the top industrial AI and robotics sessions at GTC Paris. 👀

Learn how robotics, #digitaltwins, and advanced simulation technologies are powering smarter, safer, and more efficient industrial automation. https://t.co/LyGWqzvhH2

[Premal Savla @PremalSavla]

Woo hoo! Yesterday a special engineering build of our DGX B200 system powered by Blackwell was delivered to @OpenAI. Cant wait to see what our partners build with this.

[OpenAI @OpenAI]

Look what showed up at our doorstep.

Thank you to @nvidia for delivering one of the first engineering builds of the DGX B200 to our office.

[Premal Savla @PremalSavla]

Try Qwen, a new Chinese and English LLM targeting for language, coding, mathematics, reasoning, etc. on https://t.co/pvboYAHXub. https://t.co/7hLvZemYAg

[Premal Savla @PremalSavla]

Join NVIDIA's Director of Accelerated Computing Products, Dave Salvator, in a conversation with ZKast about the NVIDIA Blackwell platform's leading performance in its MLPerf Inference debut, updates on NIM Agent Blueprints, and more. https://t.co/dfLSVrag0R

[Premal Savla @PremalSavla]

Don't miss NVIDIA's Dave Salvator, Director of Accelerated Compute, explore how to maximize AI infrastructure using our end-to-end reference architecture integrated with OCI.

Join us at #OCW24 on Sept. 11 at 9:45 AM PDT. https://t.co/6NCB7SQcRk

[Premal Savla @PremalSavla]

At #GTC24, hear from the architects of Selene and Eos about how they designed these supercomputers. Gain insights on the next generation of #NVIDIADGX architecture to power your #generativeAI applications. Register now. #DataCenter https://t.co/5QQwOWrhTY

[Premal Savla @PremalSavla]

In the latest NVIDIA AI Podcast, listen to MosaicML's mission to democratize access to #LLM’s, benefiting businesses seeking better model control, data privacy, and cost optimization. Plus - Databrick’s acquisition of #NVIDIAInception alumni @MosaicML. https://t.co/ovEdtQuO8v

[Premal Savla @PremalSavla]

Expand your robotics knowledge! Explore a free course on #IsaacSim, introducing #robotics simulation. Create realistic environments to develop and test all aspects of your robot operations. #sim2real Enroll Now. https://t.co/23ddZcTP7M

[Premal Savla @PremalSavla]

Financial firms had the second highest cybersecurity breach costs at almost $6 million last year. Learn how NVIDIA Morpheus enables #AI-based cybersecurity to identify previously undetectable threats and contain breaches faster. Join the July 11 webinar. https://t.co/MI09iDMETd

[Premal Savla @PremalSavla]

Just launched at #GTC23: #NVIDIADGX Cloud! DGX Cloud gives enterprises immediate access to a multi-node AI supercomputing service designed to train advanced models for generative #AI and other groundbreaking applications. https://t.co/EBzVquaQT6

[NVIDIA @nvidia]

Watch the #GTC23 Keynote now. Our CEO Jensen Huang shares the #AI technologies affecting every industry–and our everyday lives. https://t.co/5TS9EAl8AU

[Premal Savla @PremalSavla]

.@Amazon, @BMW, @Ericsson, @Siemens, @Lowes, and others use #NVIDIAOmniverse to plan and simulate operations, saving millions of dollars. NVIDIA-powered simulations are also helping @NASA plan the landing of its first manned mission to Mars. https://t.co/8E0dD6yhDK

[Kimberly Powell @kpowgerade]

Limited dataset availability is still one of the biggest challenges for the broad application of #AI in medicine. Learn about Rhino Health's solution to federated, secure and privacy-preserving AI for digital #health projects on Nov 10th. https://t.co/907rwEJGQI

[Premal Savla @PremalSavla]

Join us to discover how #NVIDIAInception can help your #startup build its solutions and grow faster. Registration link: https://t.co/CaWBlvOYFz #developers #entrepreneurs #AI #metropolis #innovation https://t.co/whb6HdcOD0

[Premal Savla @PremalSavla]

Today at #CloudWorld, we announced a multi-year partnership with @Oracle to help customers solve business challenges with #AI. This collaboration aims to bring our end-to-end accelerated computing stack to the Oracle Cloud Infrastructure (OCI). https://t.co/LeHOVSbdRe

[Premal Savla @PremalSavla]

Today at #CloudWorld, we announced a multi-year partnership with @Oracle to help customers solve business challenges with #AI. This collaboration aims to bring our end-to-end accelerated computing stack to the Oracle Cloud Infrastructure (OCI). https://t.co/WaBiLO9cdi

[Premal Savla @PremalSavla]

Oh this will be a great talk!!!
Marc Hamilton - Powering the Metaverse: Building Sustainable AI Factories and Digital Twins https://t.co/8IznzK3xm3