India's AI Boom: Yotta Invests $2B in Asia's Largest GPU Cluster

Yotta Data Services is making a massive $2+ billion investment to deploy one of Asia's largest AI superclusters using NVIDIA's latest Blackwell Ultra GPUs. The company's co-founder states that AI demand in India is as inevitable as mobile data and UPI adoption, driven by the country's vast digital population and talent. This infrastructure, set for a 2026 launch, aims to fill India's critical compute power gap, reducing reliance on offshore resources. The initiative supports the India AI Mission, providing compute to startups and institutions to transform India from a technology consumer to a creator.

Key Points: Yotta's $2B AI Supercluster to Power India's Population-Scale Demand

  • $2+ billion AI infrastructure investment
  • 20,736 NVIDIA Blackwell Ultra GPUs
  • Addressing India's compute power gap
  • Part of India AI Mission
  • Aiming for August 2026 launch
2 min read

AI demand inevitable at population scale like mobile data, UPI: Yotta Data Services' Co-founder

Yotta Data Services announces a $2B investment in 20,736 NVIDIA GPUs, aiming to meet India's inevitable, population-scale AI demand by 2026.

"India will require millions of GPUs... to serve our population scale AI use case. - Sunil Gupta"

New Delhi, Feb 19

Demand for artificial intelligence, for both consumers and enterprises, is inevitable at population scale, just as seen with mobile data and unified payments interface, according to Sunil Gupta, Co-founder, CEO and MD, Yotta Data Services.

Yotta Data Services has announced to deploy 20,736 liquid-cooled NVIDIA Blackwell Ultra GPUs, forming one of Asia's largest AI superclusters. This represents an investment exceeding $2 billion and is expected to go live by August 2026.

Gupta told IANS on the sidelines of the ongoing 'India AI Impact Summit 2026' here that India has the talent, ranks high on global AI skill indices, has vast and diverse datasets, multiple languages, and one of the world's largest digital populations.

"What was missing was large scale GPU infrastructure. By building supply first, we unlocked latent demand. Today, leading Indian models are trained locally, and as inferencing and enterprise adoption increase, compute availability will remain critical," he emphasised.

Yotta Data Services is empanelled with the government under the India AI Mission, which "procures compute from us on a usage basis and allocates it to startups and institutions such as Sarvam, IIT Bombay and Bhashini".

Roughly three-fourths of the GPUs deployed under the Mission come from Yotta infrastructure, he informed.

According to him, India's real gap was never demand but compute power.

"Our philosophy is -- from India, for India, and for the world," Gupta told IANS, adding that they are enabling Indian startups, hosting global models serving Indian users and serving international customers.

"India will require millions of GPUs in times to come to serve our population scale AI use case. And that is what Yotta is gearing up for," he noted.

According to the company, access to large-scale Blackwell infrastructure within the country reduces structural dependence on offshore compute and enables Indian model builders and enterprises to scale confidently.

It also allows AI products conceived in India to serve both domestic and international markets from infrastructure located within India-advancing India's ambition to evolve from a technology consumer to a technology creator.

- IANS

Share this article:

Reader Comments

P
Priya S
$2 billion is a massive investment. Hope this translates into affordable AI tools for small businesses and students, not just for big corporates. The 'for India' part needs to be real.
R
Rohit P
Finally! We have the talent and data, but were always dependent on foreign cloud for compute. This can be a game-changer for Indian startups. Building supply to unlock demand is the right strategy.
S
Sarah B
As someone working in tech, the focus on local language models (like through Bhashini) is the most exciting part. AI that understands Hindi, Tamil, Bengali etc. will have far greater reach and impact.
V
Vikram M
Good initiative, but August 2026 is a long time away in the fast-moving AI world. Hope the execution is swift and they don't face the usual infrastructure delays. The ambition is right, though.
K
Kavya N
From consumer to creator – that's the mantra we need. Jio showed it with mobile data, now hoping Yotta and others can do it with AI. Fingers crossed! 🤞

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

Leave a Comment

Minimum 50 characters 0/50