At the NVIDIA GTC AI Conference of 2024 held in California, the President and Chief Executive Officer (CEO), Jensen Huang announced its next-generation supercomputer powered by AI (artificial intelligence), the NVIDIA DGX SuperPOD. NVIDIA GB200 Grace Blackwell Superchips will power the SuperPOD. It is designed to process trillion-parameter models with constant uptime for superscale generative AI training and inference workloads. It will be faster and more powerful than its predecessor, the $40,000 H100 chip named after computer scientist Grace Hopper, which powers some of the world’s most powerful artificial intelligence models.
Blackwell is being called “the most ambitious project in Silicon Valley.” Analysts are saying that with Blackwell, NVIDIA is ushering us into the next era of supercomputers powered by AI. Amazon Web Services, Dell Technologies, Google, Meta, Microsoft, OpenAI, Oracle, Tesla, and xAI are among the top organizations that are expected to use Blackwell.
What is the NVIDIA Blackwell Platform?
GPUs, or graphics processing units, are specialized computer chips that were originally designed for rendering graphics in video games and other applications. NVIDIA is one of the leading manufacturers of GPUs, and its new Blackwell platform is specifically designed to meet the growing demand for AI computing power.
NVIDIA Blackwell is a new AI platform that is designed to power the next generation of AI products, particularly those focused on generative AI. It combines NVIDIA’s new B200 and GB200 GPUs, which are built on the Blackwell architecture and are said to be significantly faster and more energy-efficient than their predecessors. The platform also includes Confidential Computing technology to protect sensitive data and models.
Jensen Huang Net Worth 2024: Salary, House, Cars, and Other Details of Nvidia CEO
Blackwell’s revolutionary technologies include:
- World’s Most Powerful Chip: It is the most powerful chip with 208 billion transistors, made using a custom-built 4NP TSMC process.
- Second-Generation Transformer Engine: The engine boosts performance with new AI capabilities.
- Fifth-Generation NVLink: It allows for faster communication between GPUs.
- The RAS Engine: It ensures reliability and maintenance with AI diagnostics.
- Secure AI: It protects sensitive data without slowing down performance.
- A Decompression Engine that speeds up data analytics and processing.
How to Use Character AI: Features, Bypassing Filters and Other Details
When will NVIDIA Blackwell be available?
NVIDIA’s Blackwell is expected to be available at the end of the year. Amazon Web Services, Dell Technologies, Google, Meta, Microsoft, OpenAI, Oracle, Tesla, and xAI are some of the organizations that are expected to use Blackwell.
The cost of Blackwell is estimated to fall between $30,000 to $40,000. During the keynote address at the conferences, Jensen Huang said that Blackwell will be the “most successful” launch in the company’s history.
Scott Wu Net Worth: Devin AI Software Engineer, CEO of Cognition Labs
FREQUENTLY ASKED QUESTIONS
NVIDIA Blackwell is a new AI platform that is designed to power the next generation of AI products, particularly those focused on generative AI. It combines NVIDIA’s new B200 and GB200 GPUs.
The cost of Blackwell is estimated to fall between $30,000 to $40,000. It will be available in the latter half of 2024.
The NVIDIA Blackwell platform combines NVIDIA’s new B200 and GB200 GPUs, which are built on the Blackwell architecture and are said to be significantly faster and more energy-efficient than their predecessors. The platform also includes Confidential Computing technology to protect sensitive data and models.
