NxtGen Taps Diamond Cooling Tech to Deploy NVIDIA H200 GPU Servers

In a first, sovereign AI and cloud provider NxtGen will deploy NVIDIA’s H200 GPU servers cooled using ‘Diamond Cooling’ technology. The deployment, integrated with San Franciso-based Akash Systems’ patented technology, aims to increase compute density, improve energy efficiency, and reduce long-term infrastructure costs for enterprise AI workloads.

Akash Systems’ Diamond Cooling solution complements existing air and liquid cooling architectures. By eliminating thermal throttling and enabling sustained peak GPU performance, the company said the technology increases effective compute output by around 15% in high ambient temperature environments. This is particularly relevant for large-scale AI data centres operating in warmer climates such as India, where improved FLOPs per watt and enhanced power usage effectiveness are critical.

AS Rajgopal, Managing Director and CEO of NxtGen, in a statement, said the deployment strengthens India’s AI backbone by increasing usable compute per watt and per rack. “This supports structural efficiency. It allows us to deliver higher performance to customers while lowering long-term infrastructure costs.”

Diamond, the most thermally conductive material at room temperature, removes heat up to five times faster than copper. Initially proven in satellite systems, the technology now enables AI servers to sustain maximum performance even at ambient temperatures up to 50°C, compared to conventional norms of 23–29°C, according to the statement.

The deployment delivers measurable performance improvements, including lower GPU hotspot temperatures, a 15% increase in FLOPs per watt, and throttle-free operation in high-temperature environments, the company said. These outcomes translate into higher compute efficiency, improved thermal stability, and greater productivity per server.

Felix Ejeckam, Co-founder and CEO of Akash Systems, said the technology addresses two pressing constraints in AI infrastructure: energy and capital efficiency. “A 15% improvement in effective GPU compute is transformative. We are proud to partner with NxtGen on this world-first deployment.”

NxtGen’s AI platform integrates GPU architectures across NVIDIA, AMD, and Intel, offering enterprises optimised model hosting environments with sovereign data residency and compliance.