Subscribe to Our Digest
ZutaCore®, a leader in waterless, direct-to-chip, two-phase liquid cooling, today announced that its OmniTherm™ cold plate enables waterless two-phase cooling for manufacturers building servers with the NVIDIA RTX PRO 6000 Blackwell Server Edition GPUs in a single-slot PCIe form factor – supporting full-power operation in standard enterprise and AI cloud server environments.
As AI inference expands rapidly across enterprise and cloud environments, PCIe GPU servers have become a preferred platform because they are easy to deploy, scale, and integrate into existing infrastructure. As GPU power rises, however, air cooling can become a limiting system factor – restricting density, driving up fan power, which in turn increases the risk of thermal throttling during sustained workloads.
OmniTherm removes these barriers by enabling a transition to two-phase liquid cooling without introducing water inside the server. The single-slot design allows operators to increase accelerator density in standard server architectures while capturing heat into a liquid loop, reducing reliance on extreme fan speeds that can create excessive noise, waste power and cause challenging operating conditions in the data center.
“Enterprise and cloud operators want the flexibility of PCIe GPUs, but they also need density and sustained performance as power levels rise,” said My D. Troung, CTO of ZutaCore. “OmniTherm delivers waterless two-phase cooling in a single-slot form factor, helping data centers increase accelerator density while maintaining stable thermals for 24/7 AI workloads.”
Production AI workloads – especially inference – are rarely steady. They fluctuate constantly, creating thermal swings that can impact performance and reliability. ZutaCore’s two-phase approach is designed to respond rapidly to changing workloads, helping data centers maintain predictable performance under dynamic utilization.
For data center operators, reliability is as critical as raw performance. OmniTherm is designed to cool not only the GPU die, but also critical surrounding components including CPU and next-generation high-bandwidth memory. This broader cooling coverage supports long-duration, bandwidth-intensive AI workloads and helps reduce thermal stress that can impact long-term system stability.
As racks push into higher power levels, the operational cost of air cooling rises quickly, with higher fan energy consumption and growing acoustic and facility pressures. OmniTherm uses a sealed, non-conductive dielectric fluid system that enables efficient heat capture without facility water in the server. This helps reduce cooling overhead, supports improved operating conditions for data center personnel, and provides a practical path to scaling PCIe-based AI deployments.
ZutaCore is also introducing HyperCool® Cloud, a secure, cloud-native operations platform designed to help data centers manage liquid cooling infrastructure with greater visibility and lower operational risk. HyperCool Cloud provides near real-time CDU telemetry, fleet-level monitoring, and alarm-to-resolution workflows, helping operators streamline service response and strengthen uptime management as deployments scale across sites and fleets.
More information about the OmniTherm™ cold plate from ZutaCore