Video wird geladen...

Video konnte nicht geladen werden

Zur Startseite

Introducing NVFP4 — a new, AI-optimized data format that makes it faster, cheaper and greener to train AI models that can be applied to fields like healthcare, transportation and energy while preserving accuracy. Watch as Dion Harris explains our latest NVIDIA innovation:

nvidianewsroom's profile picture

NVIDIA Newsroom

52,991 subscribers

49,004 Aufrufe • vor 9 Monaten •via X (Twitter)

Gesundheit & WellnessWissenschaft & TechnologieBildung
LIVE

Anya Rossi• Live Now

Private livecam show

0 Kommentare

Keine Kommentare verfügbar

Kommentare vom Original-Post werden hier angezeigt

Ähnliche Videos

Satya Nadella predicts, - if you have AI + Quantum Computing, you may use quantum to generate synthetic data - AI can then use that data to train better models for complex fields like chemistry and physics.
0:36

Satya Nadella predicts, - if you have AI + Quantum Computing, you may use quantum to generate synthetic data - AI can then use that data to train better models for complex fields like chemistry and physics.

Haider.

78,677 Aufrufe • vor 1 Jahr

Sat down with Dion Harris, Senior Director of HPC and AI Infrastructure Solutions @ NVIDIA, to discuss how AI factories are scaling, complexities within global supply chains, and Nvidia’s latest GPU architecture, Vera Rubin. Thank you to Dion NVIDIA Data Center for taking the time!
12:18

Sat down with Dion Harris, Senior Director of HPC and AI Infrastructure Solutions @ NVIDIA, to discuss how AI factories are scaling, complexities within global supply chains, and Nvidia’s latest GPU architecture, Vera Rubin. Thank you to Dion NVIDIA Data Center for taking the time!

amit

76,014 Aufrufe • vor 4 Monaten

We are innovating the full-stack AI infrastructure platform on a one-year rhythm — pushing the boundaries of inference performance across chips, systems, and software. NVIDIA Blackwell is the most advanced AI platform ever built — continuously optimized for record-breaking performance. Groundbreaking advancements include the NVIDIA-designed NVFP4 format, enabling high inference performance and accuracy. #GTCParis at #VivaTech
1:06

We are innovating the full-stack AI infrastructure platform on a one-year rhythm — pushing the boundaries of inference performance across chips, systems, and software. NVIDIA Blackwell is the most advanced AI platform ever built — continuously optimized for record-breaking performance. Groundbreaking advancements include the NVIDIA-designed NVFP4 format, enabling high inference performance and accuracy. #GTCParis at #VivaTech

NVIDIA

80,342 Aufrufe • vor 1 Jahr

💫 Introducing our new NVIDIA DGX Spark which belongs to a new class of computers designed to build and run #AI, enabling you to prototype, fine-tune and #inference large AI models locally and seamlessly deploy to the data center or cloud. #SparkSomethingBig #GTC25
3:38

💫 Introducing our new NVIDIA DGX Spark which belongs to a new class of computers designed to build and run #AI, enabling you to prototype, fine-tune and #inference large AI models locally and seamlessly deploy to the data center or cloud. #SparkSomethingBig #GTC25

NVIDIA AI Developer

14,480 Aufrufe • vor 1 Jahr

Watch how Supermicro and NVIDIA are enabling energy-efficient, AI-optimized data centers built for tomorrow’s most demanding workloads.
0:43

Watch how Supermicro and NVIDIA are enabling energy-efficient, AI-optimized data centers built for tomorrow’s most demanding workloads.

Supermicro

80,006,902 Aufrufe • vor 10 Monaten

With our new UK data center now open, companies will be able to tap into our capacity powered by @NVIDIA Blackwell Ultra to develop new frontier models and AI solutions. Watch the short video from yesterday.
0:52

With our new UK data center now open, companies will be able to tap into our capacity powered by @NVIDIA Blackwell Ultra to develop new frontier models and AI solutions. Watch the short video from yesterday.

Nebius

53,468 Aufrufe • vor 7 Monaten

Imagine a gaming AI venturing into uncharted territories, discovering new worlds and experiences. This exploratory approach can be leveraged to generate surprisingly effective synthetic data, as our Chief AI Officer Jen D. explains. To watch previous episodes of The Greatest AI Podcast in Crypto:
0:58

Imagine a gaming AI venturing into uncharted territories, discovering new worlds and experiences. This exploratory approach can be leveraged to generate surprisingly effective synthetic data, as our Chief AI Officer Jen D. explains. To watch previous episodes of The Greatest AI Podcast in Crypto:

KIP Protocol

14,629 Aufrufe • vor 1 Jahr

Gefion, Danmark’s new #AI supercomputer, was launched today. A state-of-the-art #NVIDIADGX SuperPOD with 1528 NVIDIA H100 Tensor Core GPUs, it is ready to accelerate research in #quantum, #healthcare and #greentech. The Novo Nordisk Foundation works with Danmarks Eksport- og Investeringsfond and NVIDIA to accelerate AI driven innovation. Eviden Digital Realty Further information can be found in our press release:
2:49

Gefion, Danmark’s new #AI supercomputer, was launched today. A state-of-the-art #NVIDIADGX SuperPOD with 1528 NVIDIA H100 Tensor Core GPUs, it is ready to accelerate research in #quantum, #healthcare and #greentech. The Novo Nordisk Foundation works with Danmarks Eksport- og Investeringsfond and NVIDIA to accelerate AI driven innovation. Eviden Digital Realty Further information can be found in our press release:

Novo Nordisk Foundation

152,458 Aufrufe • vor 1 Jahr

We’re redefining what’s possible with AI. With the release of our latest model, Command A, optimized for real-world agentic and multilingual tasks, we’re demonstrating our commitment to bringing enterprises AI that goes beyond the ordinary, and offers security & efficiency. Our team has developed highly capable and efficient models that can be run on just 2 GPUs. Check out our tech report to learn more:
0:30

We’re redefining what’s possible with AI. With the release of our latest model, Command A, optimized for real-world agentic and multilingual tasks, we’re demonstrating our commitment to bringing enterprises AI that goes beyond the ordinary, and offers security & efficiency. Our team has developed highly capable and efficient models that can be run on just 2 GPUs. Check out our tech report to learn more:

Cohere

16,363 Aufrufe • vor 1 Jahr

Amjad Masad says true AGI means an AI that can enter new situations and quickly learn how to achieve goals Even the latest models still struggle with tasks outside their training data. The real breakthrough will come when AI can train itself -- like AlphaGo playing billions of games against itself
1:50

Amjad Masad says true AGI means an AI that can enter new situations and quickly learn how to achieve goals Even the latest models still struggle with tasks outside their training data. The real breakthrough will come when AI can train itself -- like AlphaGo playing billions of games against itself

Haider.

163,410 Aufrufe • vor 11 Monaten

Introducing AMD Ryzen AI Halo, a mini-PC powered by Ryzen AI Max+ that delivers desktop-class AI compute and integrated graphics for running LLMs locally. ⭐ Ready day one with the latest ROCm software, optimized AI developer workflows, and AI apps and models pre-installed.
1:00

Introducing AMD Ryzen AI Halo, a mini-PC powered by Ryzen AI Max+ that delivers desktop-class AI compute and integrated graphics for running LLMs locally. ⭐ Ready day one with the latest ROCm software, optimized AI developer workflows, and AI apps and models pre-installed.

AMD Ryzen

463,698 Aufrufe • vor 5 Monaten

Pharma might be the semiconductors of healthcare in 2026. NVIDIA-powered compute + AI design engines → faster trials, cheaper drugs, bigger margins and a wholesale sector re-rating. Watch more in this week's video:
3:57

Pharma might be the semiconductors of healthcare in 2026. NVIDIA-powered compute + AI design engines → faster trials, cheaper drugs, bigger margins and a wholesale sector re-rating. Watch more in this week's video:

Jordi Visser

42,058 Aufrufe • vor 7 Monaten

Data ownership & privacy are key to building an internet that benefits & protects its end-users AI is no different — Users’ data must be protected while interacting with AI Oasis' privacy-preserving smart contracts enable users to contribute data to AI models on their own terms
2:12

Data ownership & privacy are key to building an internet that benefits & protects its end-users AI is no different — Users’ data must be protected while interacting with AI Oasis' privacy-preserving smart contracts enable users to contribute data to AI models on their own terms

Oasis Labs

32,839 Aufrufe • vor 3 Jahren

Chonkie (Chonkie) is building the open source library for connecting your data to AI. Split unstructured data into optimized AI-ingestible chunks that boost your AI accuracy, improve app performance, and reduce token costs. Congrats on the launch, Shreyash Nigam and !
1:04

Chonkie (Chonkie) is building the open source library for connecting your data to AI. Split unstructured data into optimized AI-ingestible chunks that boost your AI accuracy, improve app performance, and reduce token costs. Congrats on the launch, Shreyash Nigam and !

Y Combinator

38,387 Aufrufe • vor 1 Jahr

Innovation is everywhere at NVIDIA GTC. 🔥 Together with NVIDIA, we are redefining enterprise AI with smarter, faster and scalable solutions that deliver ROI like never before.
0:31

Innovation is everywhere at NVIDIA GTC. 🔥 Together with NVIDIA, we are redefining enterprise AI with smarter, faster and scalable solutions that deliver ROI like never before.

Dell

74,801 Aufrufe • vor 3 Monaten

We've raised $7M to help companies build AI agents that actually learn and work. Osmosis (YC W25) is a platform for companies to fine-tune models that outperform foundation models with reinforcement learning. Better, faster, and cheaper.
1:30

We've raised $7M to help companies build AI agents that actually learn and work. Osmosis (YC W25) is a platform for companies to fine-tune models that outperform foundation models with reinforcement learning. Better, faster, and cheaper.

Kasey Zhang

1,215,152 Aufrufe • vor 8 Monaten

AI isn’t just tech anymore—it’s spreading across data, infrastructure, and energy. That shift makes diversification more important than ever. Discover how BAI and POWR can offer two different ways to play it. Learn more in our latest outlook, and join the conversation on Reddit at r/iShares.👇 #Investing #AI #ETFs
0:56

AI isn’t just tech anymore—it’s spreading across data, infrastructure, and energy. That shift makes diversification more important than ever. Discover how BAI and POWR can offer two different ways to play it. Learn more in our latest outlook, and join the conversation on Reddit at r/iShares.👇 #Investing #AI #ETFs

iShares

11,788 Aufrufe • vor 1 Monat

꧁IP꧂ encompasses all data used to fuel AI models. On Story, anyone can register open source AI models, data sets, and fine tune their models. AI cannot exist without IP, and Story makes IP programmable.
0:33

꧁IP꧂ encompasses all data used to fuel AI models. On Story, anyone can register open source AI models, data sets, and fine tune their models. AI cannot exist without IP, and Story makes IP programmable.

Story

155,404 Aufrufe • vor 1 Jahr

🌟Quilibrium’s AI Breakthrough: Encrypted Training on CPUs In her latest live stream ( - minute 14) Cassie unveiled a groundbreaking AI training method that allows models to be trained on encrypted data using CPUs while achieving performance comparable to Nvidia’s A100 GPU (blue line in the graph below). Traditionally, AI training requires expensive GPUs because matrix multiplications—the core of deep learning—are highly computational. Running these calculations on CPUs is painfully slow, often taking hours or days for even small models. The problem worsens when trying to train AI on encrypted data, as standard encryption methods add a massive computational burden. Quilibrium’s breakthrough removes this bottleneck. Instead of relying on traditional matrix multiplication, their method uses a completely different mathematical approach, allowing AI models to be trained securely and efficiently without exposing the raw data. Cassie didn’t reveal the exact technique, only hinting that it’s inspired by existing AI research and will be detailed in a future open-source AGPL-licensed paper. The key advantage? AI can now be trained at GPU speeds on standard CPUs, making privacy-preserving machine learning far more accessible. This innovation has major implications. It slashes AI infrastructure costs, allowing organizations to train powerful models without investing in expensive hardware. It also enables private AI training on personal or corporate data without revealing sensitive information, a game-changer for industries like healthcare and finance. If Quilibrium’s method delivers on its promise, it could reshape AI development, making privacy-first computing the new standard. $QUIL $wQUIL
3:42

🌟Quilibrium’s AI Breakthrough: Encrypted Training on CPUs In her latest live stream ( - minute 14) Cassie unveiled a groundbreaking AI training method that allows models to be trained on encrypted data using CPUs while achieving performance comparable to Nvidia’s A100 GPU (blue line in the graph below). Traditionally, AI training requires expensive GPUs because matrix multiplications—the core of deep learning—are highly computational. Running these calculations on CPUs is painfully slow, often taking hours or days for even small models. The problem worsens when trying to train AI on encrypted data, as standard encryption methods add a massive computational burden. Quilibrium’s breakthrough removes this bottleneck. Instead of relying on traditional matrix multiplication, their method uses a completely different mathematical approach, allowing AI models to be trained securely and efficiently without exposing the raw data. Cassie didn’t reveal the exact technique, only hinting that it’s inspired by existing AI research and will be detailed in a future open-source AGPL-licensed paper. The key advantage? AI can now be trained at GPU speeds on standard CPUs, making privacy-preserving machine learning far more accessible. This innovation has major implications. It slashes AI infrastructure costs, allowing organizations to train powerful models without investing in expensive hardware. It also enables private AI training on personal or corporate data without revealing sensitive information, a game-changer for industries like healthcare and finance. If Quilibrium’s method delivers on its promise, it could reshape AI development, making privacy-first computing the new standard. $QUIL $wQUIL

Quilibrium Community

18,887 Aufrufe • vor 1 Jahr