Frore Systems Introduces LiquidJet Nexus: Advanced Liquid Cooling for AI Data Centers

As artificial intelligence continues to drive unprecedented compute densities, managing heat has become a critical challenge for data centers. Frore Systems has announced the launch of LiquidJet Nexus, an integrated liquid cooling solution engineered specifically for the demands of next-generation AI infrastructure. This innovation addresses the escalating thermal requirements of modern AI hardware, positioning the thermal stack as a foundational layer in AI system design.

The AI Thermal Stack: A New Infrastructure Priority

With the rapid evolution of AI models and the surge in GPU and CPU power densities, the ability to efficiently dissipate heat—known as the AI Thermal Stack—is now a primary determinant of data center performance and scalability. The thermal stack encompasses the entire infrastructure responsible for extracting heat from high-performance computing components and releasing it into the environment. As AI workloads intensify, optimizing this layer is essential for maximizing compute density, system efficiency, and overall reliability.

LiquidJet Nexus: Redefining Liquid Cooling for AI Compute Trays

LiquidJet Nexus is purpose-built for liquid cooling ½ U compute trays, such as those used in NVIDIA Kyber systems. It directly cools critical components including GPUs, CPUs, DPUs, NICs, DC-DC power converters, and voltage regulators. The system integrates multiple multi-stage 3D short-loop jetchannel LiquidJet coldplates into a single, unified platform.

This innovative approach delivers several key benefits:

  • 75% higher heat transfer efficiency compared to traditional solutions
  • 8°C lower component temperatures, enhancing hardware longevity and stability
  • Elimination of hoses, connectors, and manifolds at the tray level, simplifying assembly and reducing potential failure points
  • 65% reduction in thermal stack weight, enabling higher-density compute architectures

Simplifying Data Center Cooling Architecture

Traditional liquid cooling systems often require complex assemblies of hoses, connectors, and manifolds within each compute tray. LiquidJet Nexus streamlines this process by integrating these elements into a single, system-level design. This not only reduces assembly complexity and weight but also improves reliability and supports the deployment of higher-density compute solutions.

The Three Pillars of Modern AI Infrastructure

Today’s AI systems are built on three interconnected technology layers:

  • Software Stack: AI models, agents, and applications
  • Compute Stack: GPUs, CPUs, DPUs, memory, networking, and power electronics
  • Thermal Stack: The cooling architecture that manages heat extraction and dissipation

While advancements in software and compute hardware have fueled AI’s rapid growth, the thermal stack is increasingly the limiting factor for further scaling. As Dr. Seshu Madhavapeddy, Founder and CEO of Frore Systems, notes, “The AI Thermal Stack is now a fundamental layer of AI infrastructure, and innovation in that layer will determine how far AI performance can scale.”

Thermal Innovation from Cloud to Edge

Frore Systems’ portfolio of thermal management solutions spans the full spectrum of AI deployments. LiquidJet and LiquidJet Nexus deliver advanced liquid cooling for hyperscale data centers, improving energy efficiency and compute density. For edge computing, AirJet—the world’s first solid-state active air-cooling chip—enables high-performance AI in ultra-thin, silent, dustproof, and water-resistant devices.

Together, these technologies are redefining the thermal architecture required to support AI’s rapid expansion across cloud, industrial, and consumer environments. As AI models and workloads continue to grow, innovations like LiquidJet Nexus will be essential for unlocking the next wave of performance and efficiency in data center infrastructure.