AM Intelligence orders 20,000 NVIDIA Rubin GPUs for India and Malaysia
AM Intelligence has placed two binding orders for approximately 20,000 NVIDIA Rubin GPUs, expanding its total committed capacity to around 29,000 units and close to 100 MW across India and Malaysia. The new capacity will be deployed as NVIDIA Vera Rubin NVL72 rack-scale systems, with delivery scheduled for the second quarter of 2027. This expansion follows an initial commitment for 9,000 GPUs at…
Key points
- AM Intelligence ordered 20,000 NVIDIA Rubin GPUs, raising total committed capacity to 29,000 units.
- New facilities in India and Malaysia will deliver close to 100 MW of compute by Q2 2027.
- The company plans $20 billion in additional capital expenditure for 300 MW of capacity.
The company, promoted by the founders of Greenko, stated that the new facilities will use high-throughput RDMA over Converged Ethernet networking and liquid cooling to support future AI accelerator generations. AM Intelligence aims to provide cloud providers, enterprises, and AI developers with large-scale compute closer to where workloads are developed. The company described itself as one of the early large-scale committed buyers of NVIDIA Rubin GPUs outside North America.
Beyond these orders, AM Intelligence reported a global pipeline of approximately 400 MW of Compute-as-a-Service capacity across India, the US, Europe, and Malaysia. The company plans to bring another 300 MW to market over the next 15 months, involving more than $20 billion in capital expenditure in addition to the $6 billion already committed for its initial 100 MW. Its broader strategy links AI compute with a 50 GW energy infrastructure platform in India, including solar, wind, hydro, and energy storage.
AM Intelligence Expands NVIDIA Vera Rubin AI Factory Buildout In India And Malaysia
ciol.com · 5 October 2026
Loading the full article…
This text was published by ciol.com and written by CIOL Bureau. It is reproduced here with attribution so you can read it in full; the rights remain with the publisher. Read it at the source ↗
The headline, key points and digest above were generated by Digest AI's editorial model from the linked sources. Automated summaries can contain errors: the sources are the record. Spotted a mistake? Tell us. Published by Martin K., who runs Digest AI and handles corrections.
More in Hardware & Compute
All →- User runs Qwen Flash Next Q4 on Mac Mini M5 with SSD streaming · 1 src
- Oura delays IPO as AI wearables face privacy backlash and design failures · 1 src
- Toshiba plans to double HDD production for AI data centers by FY2027 · 1 src
- SpaceX may lease AI compute to Microsoft for $150B deal · 2 src
- Strata runtime doubles Qwen3.8-Flash-Next speed on RTX 4070 · 1 src
Comments
via GitHub Discussions