Hiranandani-Backed Yotta Data Services Announces $2 Billion+ NVIDIA Blackwell Ultra GPU Supercluster, One of Asia's Largest AI Infrastructure Deployments

Yotta Deploys 20,736 NVIDIA Blackwell Ultra GPUs in Asia's Largest AI Supercluster

Yotta Data Services announced it will deploy 20,736 liquid-cooled NVIDIA Blackwell Ultra GPUs, forming one of Asia's largest AI superclusters. The infrastructure is expected to go live by August 2026. The $2 billion-plus investment will establish one of Asia's largest AI superclusters, marking a major expansion of compute capacity for India's emerging AI infrastructure ecosystem.

The deployment positions India among a select group of geographies capable of hosting frontier-scale AI infrastructure. Yotta Data Services is the data centre and cloud division of the Hiranandani Group, founded in 2019 to develop integrated data centre parks across India.

Greater Noida Campus and Facility Scope

The supercluster will be deployed at Yotta's 60 MW D2 hyperscale data centre at its Greater Noida campus, which is scalable to 250 MW. Additional capacity will be supported by Yotta's Navi Mumbai campus, scalable to 2 GW. Yotta Greater Noida Data Center Park is provisioned for 6 interconnected data center buildings.

The facility is part of a larger ecosystem. G1 is Yotta's fifth data centre in India, with two others located in Navi Mumbai and Greater Noida as part of hyper scale campuses. Yotta D1 in Greater Noida is North India's hyperscale Tier III Certified data centre with low-latency connectivity, resilient power, and enterprise-grade colocation infrastructure.

Technical Architecture and Specifications

The supercluster is built on Nvidia reference architecture and integrates 800 Gbps Nvidia Quantum-X800 InfiniBand networking, advanced liquid-cooling systems, and over 40 petabytes of high-performance parallel file-system storage. The platform is designed to support trillion-parameter foundation model training and high-throughput inference workloads capable of handling multi-million simultaneous prompts.

NVIDIA shipped the B300 (officially "Blackwell Ultra") in January 2026. It's the company's most powerful single GPU to date, with 288 GB of HBM3e memory, 8 TB/s bandwidth, and 15 petaFLOPS of dense FP4 compute per chip. New Blackwell Ultra Tensor Cores deliver 1.5x more AI compute FLOPS compared to Blackwell GPUs or 70x more AI FLOPS for GB300 NVL72 compared to HGX H100.

NVIDIA DGX Cloud Partnership

Separately, Yotta has signed a four-year commercial engagement worth over $1 billion with Nvidia. Under this, Nvidia will establish one of APAC's largest Nvidia DGX Cloud cluster within Yotta's HGX B300 Blackwell Ultra supercluster. Nvidia DGX Cloud has been utilising Yotta's GPU infrastructure over the past year, and the expanded deployment extends that relationship.

Scale and Roadmap

Yotta currently operates over 10,000 NVIDIA GPUs live in production, with another 8,000 NVIDIA GPUs going live within the next quarter, followed by the deployment of 20,736 Blackwell Ultra GPUs by August 2026. The company has outlined a roadmap to scale beyond 80,000 NVIDIA GPUs by FY27, supported by phased infrastructure expansion and long-term capacity planning.

Yotta said its platform can scale beyond one million GPUs within the next three to five years as India's AI ecosystem accelerates.

Hiranandani Group's Data Centre Division

Darshan Hiranandani is the Chairman and Co-founder of Yotta Data Services. He also serves as CEO of the Hiranandani Group, a leading real estate and infrastructure development company. He built and operated the highest number of (>19) 3rd party Data Centers in India including some of the largest hyperscale DCs meeting the stringent quality demands of global OTT players.

The Hiranandani Group, which owns over 1800 acres of land across India and is a leading construction company, is now making a foray into Data Centre Parks, and their new business venture is called Yotta Infrastructure. Yotta represents a strategic diversification of the group's infrastructure portfolio, leveraging the organisation's existing real estate and operational capabilities to build India's hyperscale compute infrastructure.