AM Group partners OXMIQ Labs for 2 GW AI compute plan

AM Group (AMG) has partnered with OXMIQ Labs to provide data centre and system infrastructure advisory for a planned 2 GW artificial intelligence (AI) compute capability by 2030, starting with a 1 GW AI Compute Hub in Uttar Pradesh. The engagement will be led through AMI Labs, a business division of AMG.

Project background
AMG has committed USD 25 billion to develop an AI infrastructure hub in the Greater Noida region of Uttar Pradesh. AMI Labs, the group’s technology division and part of AMG, the parent company of Greenko Group, operates with support from 50 GW of renewable energy capacity across solar, wind, and hydro assets, along with 100 GWh of intelligent energy storage. AMG currently supplies around 2% of India’s total power, with generation that is owned, operated, and carbon-free, and priced at 50–70% below conventional data-centre power costs.

Infrastructure scope
The Noida facility is planned as a vertically integrated platform combining carbon-free power generation, data-centre infrastructure, high-performance accelerators, software systems, and application layers. It will also offer flexible consumption models, including AI Pods-as-a-Service and Tokens-as-a-Service. Phase 1 of the Compute Hub is under execution, with initial compute capacity targeted by the end of 2027.

OXMIQ’s role
OXMIQ will act as the architecture and engineering partner for the compute platform. Its scope includes systems architecture, hardware roadmap planning, and supply-chain strategy. The company will work across the compute stack, covering graphics processing unit (GPU) architecture, advanced packaging, rack-scale systems, high-performance interconnects, and workload orchestration software. The collaboration also focuses on system optimisation, infrastructure design, and modular deployment to support rapid scaling.

Integrated design approach
The project adopts a fully integrated design across energy generation, data-centre infrastructure, liquid cooling systems, interconnect topology, accelerator selection, and workload orchestration. This approach aims to deliver lower-cost AI compute at gigawatt scale.

Market context
The development aligns with India’s position as the world’s second-largest market for AI usage and token consumption, supported by growth in the developer ecosystem, expansion of the digital economy, and increasing enterprise adoption of AI.

Anil Chalamalasetty, Group Chairman of AMG, stated that OXMIQ’s hardware and systems expertise would support the development of globally competitive infrastructure and a full-stack AI compute platform. Raja Koduri, Founder and Chief Executive Officer (CEO) of OXMIQ Labs, said AMI Labs has addressed a key constraint in large-scale AI infrastructure, access to reliable, carbon-free power, and added that early-stage collaboration would align system design with AI workload requirements and cost structures.

The featured photograph is for representation only.