Nvidia unveils Blackwell GPU offering five times the performance for AI inference

Blackwell GPU contains 208 billion transistors and is paired with 192GB of HBM3e memory

clock • 3 min read
Jensen Huang
Image:

Jensen Huang

Nvidia has unveiled its latest B200 Blackwell GPU that, it claims, is twice as powerful as current GPUs for training AI models, and offers five times the performance at inference, the speed with which AI models respond to queries.

The Blackwell architecture contains 208 billion transistors and will be manufactured on TSMC's four nanometre 4NP process node, a refined version of the 4N process used to produce existing Nvidia H...

To continue reading this article...

Join Computing

  • Unlimited access to real-time news, analysis and opinion from the technology industry
  • Receive important and breaking news in our daily newsletter
  • Be the first to hear about our events and awards programmes
  • Join live member only interviews with IT leaders at the ‘IT Lounge’; your chance to ask your burning tech questions and have them answered
  • Access to the Computing Delta hub providing market intelligence and research
  • Receive our members-only newsletter with exclusive opinion pieces from senior IT Leaders

Join now

 

Already a Computing member?

Login

You may also like
Nvidia outlines AI strategies for GPUs and Kubernetes

Chips and Components

At the KubeCon event in Paris this week, Nvidia engineers spoke of challenges and solutions for accelerating AI workloads using GPUs with Kubernetes.

clock 22 March 2024 • 3 min read
Don't encourage kids to code, says Nvidia CEO

Skills

AI threatens the future of coding

clock 29 February 2024 • 3 min read
Nvida revenues rise 265% on AI chips

Finance and Reporting

Looks set to make Nvidia more valuable than Amazon and Alphabet

clock 22 February 2024 • 2 min read

Sign up to our newsletter

The best news, stories, features and photos from the day in one perfectly formed email.

More on Chips and Components

Microsoft claims 'most reliable logical qubits on record'

Microsoft claims 'most reliable logical qubits on record'

We're entering the age of resilient quantum computing, say Microsoft and Quantiniuum

clock 04 April 2024 • 2 min read
Pragmatic Semiconductor opens wafer fab in Durham

Pragmatic Semiconductor opens wafer fab in Durham

UK's first 300mm facility

Tom Allen
clock 28 March 2024 • 2 min read
Nvidia outlines AI strategies for GPUs and Kubernetes

Nvidia outlines AI strategies for GPUs and Kubernetes

At the KubeCon event in Paris this week, Nvidia engineers spoke of challenges and solutions for accelerating AI workloads using GPUs with Kubernetes.

John Leonard
clock 22 March 2024 • 3 min read