Iceotope’s design circulates a small volume of sealed, non-conductive dielectric coolant past heat-producing components throughout a server—not just the CPUs and GPUs. The company calls this “direct-to-everything” cooling. Its KUL AI platform, launched on December 12, 2024, applies the approach to an 8-GPU server.
How Iceotope’s “direct-to-everything” cooling works
Iceotope places dielectric coolant inside the server chassis and circulates it through an internal manifold. The fluid passes the processors, memory, storage and power-supply components, absorbing heat through forced convection. It then carries that heat to a liquid-to-liquid heat exchanger, which transfers it into a separate Technology Cooling System (TCS) loop.
This is not simply a conventional server with cold plates on its main processors. The claimed distinction is the breadth of component coverage: the sealed fluid circuit is designed to cool the system’s heat-producing hardware, while the TCS loop handles heat transfer out of the chassis.
Is KUL AI immersion cooling or direct-to-chip?
Iceotope positions the architecture between the usual descriptions of immersion and direct-to-chip cooling. It uses a sealed dielectric fluid within the chassis, a feature associated with immersion approaches, but circulates that fluid through an internal system to remove heat from components. It is therefore more precise to call it Iceotope’s own precision liquid-cooling or “direct-to-everything” architecture than to label KUL AI simply an immersion tank or a CPU/GPU cold-plate system.
Free tools Windows power users keep installed
One-click scans. No signup required.
#1 Best Overall
- Engineered for High-Performance Computing: Supports E-ATX motherboards for multi-GPU setups and top-tier hardware, making it a solid foundation for rendering, AI training, and virtualization servers
- Ready for 360mm AIO Liquid Cooling: Directly supports a 360mm radiator for extreme CPU cooling, enabling sustained performance under heavy computational loads without modification
- Triple 120mm PWM Fans for Effective Airflow: Three pre-installed, quiet PWM fans work with the AIO to ensure optimal airflow, keeping GPUs and other critical components cool and stable
- Front-Access 10Gbps USB-C for Fast Transfers: A front-panel USB 3.2 Gen Type-C port delivers up to 10 Gbps transfer speeds, drastically improving efficiency when working with large files
- Versatile Storage Configuration: Accommodates up to 2 x 3.5-inch hard disk drives and 4 x 2.5-inch solid state drives, providing flexible storage options for your server needs
| Question | Iceotope’s approach |
|---|---|
| What is cooled? | Processors, memory, storage and power-supply components, according to Iceotope. |
| What fluid is inside the server? | A small amount of non-conductive dielectric coolant, sealed inside the chassis. |
| How does heat leave the chassis? | Through a liquid-to-liquid heat exchanger into a separate TCS loop. |
| What product applies the design? | KUL AI, an 8-GPU precision-liquid-cooled server platform launched December 12, 2024. |
Why the operating temperature matters
Iceotope says the design supports inlet coolant temperatures above 55°C. That is relevant because a warmer water loop can make heat reuse easier: the facility may be able to use the captured heat without cooling the loop as aggressively first. The claim describes a supported inlet temperature, not a guarantee that every installation will operate at that temperature or recover useful heat; outcomes depend on the facility and its heat-reuse setup.
Cooling more than the primary processors may also reduce dependence on server fans and help address the thermal load of dense AI and high-performance computing systems. Iceotope presents KUL AI for data centers, edge sites and other settings where thermal management and uptime matter. Actual rack density, retrofit effort and service procedures are deployment-specific; the figures provided for the design do not establish those outcomes for every site.
Rank #2
- All in One Liquid Cooler
- Supports AMD sWRX8/sTRX4/TR4/SP3
- Supports up to 280 Watts Heat Dissipation For 1U server and up
- Shin-Etsu 7762 or Equivalent Pre-Printed Thermal Compound
What Iceotope claims about energy, water and emissions
Iceotope’s current technology page reports the following reductions per kW of ITE (information technology equipment) power. These are vendor-reported figures; the available material does not provide an independent audit methodology or enough detail to establish the comparison conditions.
| Reported measure | Iceotope’s claim | Qualification |
|---|---|---|
| Power usage | 40% less | Per kW of ITE power; vendor-reported. |
| Water usage | 100% less | Per kW of ITE power; vendor-reported. |
| Carbon emissions | 40% lower | Per kW of ITE power; vendor-reported. |
| Cooling usage | 84% less | Per kW of ITE power; vendor-reported. |
Those percentages should be read as company claims, not as independently verified savings that every buyer can expect. The stated basis is per kW of ITE power, but the published figures here do not specify the comparator, test conditions, facility boundary or whether the reductions apply to a particular configuration. In particular, the 100% water-usage claim should not be taken to mean that a data center has no water use of any kind.
Rank #3
- LGA 115X, LGA1700, LGA2011/2066 Square ILM AM2/AM3/AM4/AM5/FM1/FM2
- 1U Server
- 2 Ball Bearing
- At Duty Cycle 100%: 53.96 dBA
- At Duty Cycle 0~ 20%: 0.72 W At Duty Cycle 50%: 1.56 W At Duty Cycle 100%: 8.40 W
What the 2026 patent announcement adds
In April 2026, Iceotope said its portfolio included 200 granted and pending patents spanning chassis design, dielectric-fluid use and rack-scale thermal management. The company’s patent-milestone announcement also described chip-level cooling of up to 1500W and up to 40% lower energy use versus traditional air cooling. These are Iceotope’s stated patent-announcement figures; they are not, by themselves, independently validated performance results for every KUL AI deployment.
Independent reader supportYour contribution helps us test, update, and keep practical guides available for everyone.What to assess before considering a deployment
The architecture addresses server heat removal, but adopting it is an infrastructure decision. Buyers evaluating it should establish how the sealed chassis circuit connects to the site’s cooling system and what changes their deployment requires.
Rank #4
- Dynatron L3 Aio Enclosed 1u Server Liquid Cooler Intel/amd Socket Brown Box
- Coverage: Confirm which components in the target server are cooled by the internal fluid circuit, and whether any remaining heat loads rely on air cooling.
- Facility loop: Determine the required TCS interface, supported inlet conditions and compatibility with the site’s existing cooling plant.
- Heat reuse: Check whether a suitable use for recovered heat exists at the proposed operating temperature; a warm loop alone does not create a heat-reuse system.
- Service and retrofit: Get deployment-specific guidance on maintenance, access to components and whether existing servers or racks can be adapted. The cited claims do not quantify retrofit requirements or serviceability.
- Performance evidence: Ask for the comparison baseline, measurement boundary and operating conditions behind energy, water and emissions estimates before using those figures in a site business case.
Iceotope also announced a $26 million Series B in May 2026 to expand product and engineering development, patents and ecosystem partnerships. That indicates a company investment plan, not a separate measure of KUL AI’s cooling performance.
Quick Recap
Best Value
- Liquid cooling radiators: Support up to 360mm
- M/B size: EATX/ATX/MicroATX/Mini-ITX
- Drive Bays: 2*3.5 (internal)
- Expansion Slots: 8xslots PCI/PCIE full height
- Sliding rail: Not support, suggest to use rack shelf
Product prices and availability are accurate as of the date/time indicated and are subject to change. Any price and availability information displayed on Amazon at the time of purchase will apply.
