Nvidia Strengthens AI Dominance with New ‘Super Chip’ Blackwell
Nvidia announced a new generation of AI chips and software for running AI models.
Topics
News
- Core42 and Qualcomm Launch Compass 2.0 to Boost AI Efficiency and Accessibility
- Data Debt Emerges as Core Business Issue for Enterprises
- Are Organizations Investing in AI Achieving Positive ROI?
- Are Increasing Breaches Due to the Cyber Skills Gap?
- What Are the Challenges Firms in the UAE and Saudi Arabia Face in GenAI Implementation?
- Enterprises See the Promise of Generative AI But Lack the Guardrails to Mitigate Operational Risks
[Image source: Krishna Prasad/MITSMR Middle East]
While companies and software makers, for example, are still scrambling to get their hands on the current Hopper H100s and similar chips, Nvidia announced a new generation of AI chips and software for running AI models, the Blackwell B200.
The announcement comes as the chipmaker seeks to solidify its position as the go-to supplier for AI companies.
Nvidia’s high-end server GPUs are essential for training and deploying large AI models. Companies like Microsoft and Meta have spent billions of dollars buying the chips.
The Blackwell chip will replace the Hopper, which helped boost huge sales for Nvidia.
Nvidia chief executive Jensen Huang said the latest chips can be installed by simply sliding out the Hopper units from existing infrastructure.
“Hopper is fantastic, but we need bigger GPUs,” he said.
Huang also declared the newest “super chip” would be “the most successful product launch in our history.”
Nvidia says that the company is becoming less of a platform provider, like Microsoft or Apple, on which other companies can build software.
“Blackwell’s not a chip, it’s the name of a platform,” Huang said.
Every two years, Nvidia updates its GPU architecture, unlocking a big jump in performance. Many AI models released over the past year were trained on the company’s Hopper architecture.
Nvidia says Blackwell-based processors, like the GB200, offer a huge performance upgrade for AI companies, with 20 petaflops in AI performance versus 4 petaflops for the H100. The additional processing power will enable AI companies to train bigger and more intricate models, Nvidia said.
The chip includes what Nvidia calls a “transformer engine specifically built to run transformers-based AI, one of the core technologies underpinning ChatGPT.
The Blackwell GPU is large and combines two separately manufactured dies into one chip manufactured by TSMC. It will also be available as an entire server called the GB200 NVLink 2, combining 72 Blackwell GPUs and other Nvidia parts designed to train AI models.
To become leaders in AI and turbocharge their economies, Saudi Arabia and the UAE have been big buyers of the high-performance Nvidia chips crucial for building AI software.
Last year, there were reports that Saudi Arabia had bought around 3,000 of Nvidia’s H100 chips, as its LLM is being developed by the Provable Responsible AI and Data Analytics lab at King Abdullah University of Science and Technology. The UAE has also secured access to thousands of Nvidia chips and developed its open-source large language model, Falcon, at the state-owned Technology Innovation Institute in Abu Dhabi. The UAE’s Falcon model was trained on 384 A100 chips.
Last year, OpenAI chief executive Sam Altman praised Abu Dhabi’s foresight in recognizing the importance of AI during a visit to the region. He said the Gulf region could “play a central role in this global conversation” around the emerging technology and its regulation.
At the NextTech Summit, the Middle East’s foremost summit focusing on emerging technologies, global experts, MIT professors, industry leaders, policymakers, and futurists will discuss emerging technologies, such as Responsible AI, Quantum Computing, Human-Machine Collaboration, among many other technologies and their immense potential. Click here to register.
