AM Intelligence (AMI), the AI infrastructure platform set up by the promoters of Greenko, announced on 5 October 2026 two more firm orders for about 20,000 NVIDIA Rubin GPUs, to be deployed in India and Malaysia.
The new orders add about 70 MW of compute and are due for delivery in the second quarter of 2027. AMI's committed capacity is now about 29,000 Rubin GPUs and close to 100 MW.
Its first order, announced on 25 August 2026, was for 9,000 Rubin GPUs (30 MW) for its first AI factory in Hyderabad.
AMI plans another 300 MW in the next 15 months at a capital cost of over USD 20 billion, on top of USD 6 billion committed for the first 100 MW. Its stated goal is 1 GW of Compute-as-a-Service capacity.
A company builds data centres full of GPUs and rents the computing power to AI developers, who pay for use and do not buy chips. A GPU handles many calculations in parallel, which is what training AI models needs.
Simple Analogy: Like buying electricity from a power plant without building your own.
AI infrastructure platform set up by the promoters of Greenko, the Hyderabad-based renewable energy company; sells Compute-as-a-Service
American chip designer founded in 1993; the leading supplier of GPUs for AI; maker of the Vera Rubin platform
Build public AI computing capacity and support Indian AI start-ups, datasets and skills
Key: Approved by the Union Cabinet in March 2024 with an outlay of Rs 10,371.92 crore; implemented by the Ministry of Electronics and Information Technology; its compute pillar targets 10,000 or more GPUs. AMI's project is private and separate from it
General Awareness > Science and Technology > Computers and AI
Graphics processing unit, a chip that performs many calculations in parallel and is used to train and run AI models
A data centre built mainly to train and run AI models at scale