Topics
Latest
AI
Amazon
Image Credits:Nvidia
Apps
Biotech & Health
Climate
Image Credits:Nvidia
Cloud Computing
Commerce
Crypto
It is not a huge surprise, perhaps, that the Nvidia CEO drew parallels to a concert. The venue was, in a word, very concert-y.Image Credits:TechCrunch / Haje Kamps
Enterprise
EVs
Fintech
fundraise
Gadgets
Gaming
Government & Policy
computer hardware
Layoffs
Media & Entertainment
Meta
Microsoft
seclusion
Robotics
Security
Social
Space
Startups
TikTok
Transportation
Venture
More from TechCrunch
Events
Startup Battlefield
StrictlyVC
newssheet
Podcasts
picture
Partner Content
TechCrunch Brand Studio
Crunchboard
Contact Us
“ I go for you realize this is not a concert , ” said Nvidia chairwoman Jensen Huang to an consultation so large , it filled up the SAP Center in San Jose . This is how he introduced what is perhaps the complete opposite of a concert : the company ’s GTC issue . “ You have make it at a developers group discussion . There will be a lot of science describing algorithms , figurer architecture , mathematics . I smell out a very heavy weight in the elbow room ; all of a sudden , you ’re in the wrong plaza . ”
It may not have been a sway concert , but the leather - jacket - wear 61 - yr - old CEO ofthe world ’s third - most - worthful companyby market place cap certainly had a fair number of fan in the audience . The ship’s company launched in 1993 with a mission to crowd general computing past its limits . “ Accelerated computing ” became the razz cry for Nvidia : Would n’t it be great to make chips and board that were specialized , rather than for a general use ? Nvidia chips give art - thirsty gamers the tool they take to bet games in higher resolution , with higher quality and gamy frame rates .
Monday ’s tonic was , in a direction , a return to the caller ’s original mission . “ I want to show you the soul of Nvidia , the soul of our company , at the point of intersection of computer graphics , natural philosophy and contrived intelligence , all intersect inside a computer . ”
Then , for the next two hour , Huang did a rarified thing : He nerded out . firmly . Anyone who came to the keynote expecting him to pull a Tim Cook , with a slick , consultation - focused tonic , was bound to be disappoint . Overall , the keynote was tech - intemperate , acronym - riddled , and unapologetically a developer conference .
We need bigger GPUs
computer graphic processing unit of measurement ( GPUs ) are where Nvidia got its start . If you ’ve ever built a computing machine , you ’re probably thinking of a nontextual matter scorecard that hold out in a PCI slot . That is where the journey started , but we ’ve come a farsighted way since then .
The company announced its mark - new Blackwell program , which is an infrangible monster . Huang says that the core of the processor was “ drive the limits of natural philosophy of how big a chip could be . ” It combines the power of two chips , offering speeds of 10 Tbps .
Join us at TechCrunch Sessions: AI
Exhibit at TechCrunch Sessions: AI
“ I ’m holding around $ 10 billion worth of equipment here , ” Huang say , holding up a prototype of Blackwell . “ The next one will be $ 5 billion . Luckily for you all , it experience cheaper from there . ” Putting a cluster of these crisp together can crank out some really telling power .
The old coevals of AI - optimize GPU was called Hopper . Blackwell is between 2 and 30 time faster , look on how you mensurate it . Huang explained that it took 8,000 GPUs , 15 megawatts and 90 daytime to make the GPT - MoE-1.8 T manikin . With the new system , you could use just 2,000 GPUs and use 25 % of the superpower .
These GPUs are pushing a rattling amount of data around — which is a very good segue into another topic Huang talked about .
What’s next
Nvidia wheel out anew exercise set of toolsfor auto manufacturer working on self - driving motorcar . The ship’s company was alreadya major player in robotics , but it double down with new dick for roboticiststo make their robots smarter .
Huang kept repeating the phrase“AI factory , ” rather than datum center . “There ’s a novel Industrial Revolution encounter in these [ waiter ] room : I call them AI factories , ” Huang said .
The society also introducedNvidia NIM , a software program aimed at simplify the deployment of AI framework . NIM leverages Nvidia ’s hardware as a creation and aims to speed companies ’ AI initiatives by providing an ecosystem of AI - ready containers . It supports models from various sources , including Nvidia , Google and Hugging Face , and integrates with platforms like Amazon SageMaker and Microsoft Azure AI . NIM will dilate its capability over time , including tools for generative AI chatbots .
“ Anything you could digitalise : So long as there is some structure where we can employ some design , intend we can learn the patterns , ” Huang said . “ And if we can learn the patterns , we can understand the signification . When we understand the meaning , we can mother it as well . And here we are , in the reproductive AI revolution . ”
Catch up on Nvidia ’s GTC 2024 :
Update : This military post was updated to admit unexampled information and a TV of the keynote .
