NewKerala.com

South Korea's AI Boost: 10,000 Nvidia GPUs for Startups and SMEs

South Korea is taking a major step to boost its AI capabilities. The government plans to distribute 10,000 powerful Nvidia graphics processors to small businesses, startups, and research projects starting in February. This is the first part of a much bigger plan to get 50,000 of these chips by 2030, costing nearly a billion dollars. The move is all about helping local companies build their own AI models and stay competitive.

Key Points: South Korea Allocates 10,000 Nvidia GPUs to SMEs and AI Projects

  • Government to allocate GPUs starting in February, with applications open until Jan. 28
  • First phase of a $947M plan to acquire 50,000 Nvidia GPUs by 2030
  • Aims to foster a robust domestic AI ecosystem and develop homegrown AI models
  • Part of a larger deal with Nvidia to deploy 260,000 GPUs nationwide involving major corporations
2 min read

S. Korea to allocate 10,000 Nvidia GPUs to SMEs, startups, AI projects

South Korea begins allocating 10,000 Nvidia GPUs to SMEs, startups, and AI projects in February, part of a $947M investment to build a sovereign AI platform.

"The GPUs will be configured as a large-scale cluster to support high-speed computing for AI model training and inference. - Government Announcement"

Seoul, Dec 18

The government here said on Thursday that it plans to allocate some 10,000 graphics processing units (GPUs) procured from Nvidia to small and mid-sized enterprises (SMEs), startups and academic and state artificial intelligence (AI) projects starting in February.

The plan was unveiled as part of the government's broader strategy to foster a robust domestic AI ecosystem at a meeting of science and technology-related ministers chaired by Science Minister Bae Kyung-hoon, the government said.

The government recently spent 1.4 trillion won (US$947.2 million) to purchase some 10,000 GPUs from Nvidia, marking the first phase of a plan to acquire 50,000 units by 2030, reports Yonhap news agency.

The GPUs will be configured as a large-scale cluster to support high-speed computing for AI model training and inference.

Applications for AI development projects from industry, academia and research institutions will be accepted online until Jan. 28, with support of up to 256 H200 GPUs or 128 B200 GPUs per project.

Some 6,000 B200 GPUs, scheduled for later delivery, will be allocated to a nationwide initiative to develop homegrown AI foundation models, the government added.

The allocation follows Nvidia Chief Executive Officer (CEO) Jensen Huang's announcement in late October that the company would deploy 260,000 GPUs to South Korea in partnership with the government and major companies, including Samsung Electronics Co., to build large-scale AI factories in the country.

Under the plan, the government will use some 50,000 GPUs to build a national sovereign AI platform, while Samsung Electronics, SK Group and Hyundai Motor Group are each set to receive 50,000 units. Naver Corp. will receive 60,000.

Earlier this month, the South Korean government received the first batch of Nvidia graphics processing units (GPUs) as part of the company's plan to deploy 260,000 units to the country, government sources said Monday.

The government executed 1.4 trillion won (US$952.7 million) in budget to bring in 13,000 GPUs from Nvidia, according to the Ministry of Science and ICT and other sources.

- IANS

Share this article:

Reader Comments

P
Priya S
Wow, 10,000 GPUs just for SMEs and startups? That's the kind of support that can create unicorns. Our Indian startups are often scrambling for compute resources. Hope our IT ministry is taking notes. The focus on homegrown foundation models is especially crucial.
R
Rohit P
The scale is mind-boggling - 260,000 GPUs in total! Samsung, SK, Hyundai all getting 50k each. This will supercharge their industries. We need our Tata, Reliance, and Infosys to make similar bold moves, but government partnership is key. The 'AI factory' concept is the future.
S
Sarah B
As someone working in AI research here in India, access to high-end GPUs is our biggest bottleneck. A program like this, where academics can apply for up to 256 H200s, would be a game-changer. We're years behind without this kind of national compute infrastructure.
V
Vikram M
It's a smart strategic move. They're not just buying chips; they're building a complete ecosystem with a sovereign platform. Respectfully, I feel our Digital India and AI missions talk a lot but need to move faster on tangible hardware procurement. The global AI race is accelerating.
K
Kavya N
The budget is huge - nearly $1 billion just for this phase. But it's an investment in the future. If India wants to be a Vishwaguru in tech, we need to allocate funds for such capital-intensive tech infra. Hope to see similar announcements in the next budget session! 🤞

We welcome thoughtful discussions from our readers. Please keep comments respectful and on-topic.

Leave a Comment

Minimum 50 characters 0/50