We Keep it Cool at Scale.
Solving Power, Heat and Scale Challenges for AI
Flex, the global partner of choice for advanced manufacturing, data center IT and power infrastructure solutions from the grid to the chip, offers customizable server designs engineered to the specific power, compute, and cooling demands of hyperscalers and enterprise customers. Flex’s end-to-end infrastructure solutions integrate compute, power distribution, and JetCool’s liquid cooling technology into a complete rack-level solution, delivering integrated cooling and server technology that supports the most demanding AI and high-performance computing (HPC) workloads with unmatched efficiency and scalability.
Flex’s portfolio of services and power products cover over 80% of the data center
The JetCool and Flex Advantage
By combining Flex’s high-volume manufacturing expertise with JetCool’s patented microconvective cooling technology, hyperscalers can deploy advanced AI and HPC systems with reduced time-to-market. JetCool’s microjet technology leverages impinging microjets to achieve a significantly lower thermal resistance compared to conventional microchannel cooling. This innovation drives superior heat extraction at the die level, accelerating thermal management optimization and enabling higher performance densities. The integration of microjet cooling within Flex’s system designs reduces temperature gradients and supports a more uniform thermal distribution across complex silicon architectures.
Solving Power, Heat and Scale Challenges for AI
Flex, the global partner of choice for advanced manufacturing, data center IT and power infrastructure solutions from the grid to the chip, offers customizable server designs engineered to the specific power, compute, and cooling demands of hyperscalers and enterprise customers. Flex’s end-to-end infrastructure solutions integrate compute, power distribution, and JetCool’s liquid cooling technology into a complete rack-level solution, delivering integrated cooling and server technology that supports the most demanding AI and high-performance computing (HPC) workloads with unmatched efficiency and scalability.
Flex’s portfolio of services and power products cover over 80% of the data center
The JetCool and Flex Advantage
By combining Flex’s high-volume manufacturing expertise with JetCool’s patented microconvective cooling technology, hyperscalers can deploy advanced AI and HPC systems with reduced time-to-market. JetCool’s microjet technology leverages impinging microjets to achieve a significantly lower thermal resistance compared to conventional microchannel cooling. This innovation drives superior heat extraction at the die level, accelerating thermal management optimization and enabling higher performance densities.
Scalable, Vertically-Integrated Liquid-Cooled Racks for AI
Accelerating deployment at scale
With the increasing complexity and scale of computing in the AI era, integrated system rack designs that bring together compute, storage, network, power, and cooling technologies are imperative. Balancing rack standardization and customization, OCP introduced the Open Rack v3 (ORv3) specification to drive efficiency into the design and delivery of integrated rack solutions that meet several hyperscaler design points. Innovative high-density racks deliver breakthroughs in compute performance per square foot, increasing data center capacity return on investment (ROI) while supporting scalability.
In partnership with Flex, leading hyperscalers are deploying customized ORv3-based rack designs at scale. They are also tapping Flex to manage vertical integration of data center rack solutions, from the fabrication of sheet metal frames and enclosures to the design and manufacture of servers, storage, racks, cabling, switches, busbars, power shelves, battery backup, and liquid cooling systems.
Solving Power, Heat and Scale Challenges for AI
Flex, the global partner of choice for advanced manufacturing, data center IT and power infrastructure solutions from the grid to the chip, offers customizable server designs engineered to the specific power, compute, and cooling demands of hyperscalers and enterprise customers. Flex’s end-to-end infrastructure solutions integrate compute, power distribution, and JetCool’s liquid cooling technology into a complete rack-level solution, delivering integrated cooling and server technology that supports the most demanding AI and high-performance computing (HPC) workloads with unmatched efficiency and scalability.
By combining Flex’s high-volume manufacturing expertise with JetCool’s patented microconvective cooling technology, hyperscalers can deploy advanced AI and HPC systems with reduced time-to-market. JetCool’s microjet technology leverages impinging microjets to achieve a significantly lower thermal resistance compared to conventional microchannel cooling. This innovation drives superior heat extraction at the die level, accelerating thermal management optimization and enabling higher performance densities. The integration of microjet cooling within Flex’s system designs reduces temperature gradients and supports a more uniform thermal distribution across complex silicon architectures.
Scalable, Vertically-Integrated Liquid-Cooled Racks for AI
Accelerating deployment at scale
In partnership with Flex, leading hyperscalers are deploying customized ORv3-based rack designs at scale. They are also tapping Flex to manage vertical integration of data center rack solutions, from the fabrication of sheet metal frames and enclosures to the design and manufacture of servers, storage, racks, cabling, switches, busbars, power shelves, battery backup, and liquid cooling systems.
Explore JetCool's Other Products
SmartPlate System for Dell PowerEdge Servers
SmartPlate System is a chassis-level, closed-loop, liquid-assisted air cooling solution. This system is outfitted with two enhanced cold plates that utilize our patented microconvective cooling® technology, a pump, and radiator. The cold plates directly target the processor’s hotspots, enabling the system to be more efficient and able to cool the industry’s highest power processors.
- Save an average of 15% on server power savings overnight
- Supports 850W in 1U configurations and 1,200W in 2U configurations
- Maximize compute per floor tile and minimize facility floorspace with no piping or plumbing
Direct-to-Chip Single-Phase Liquid Cold Plates for HPC
JetCool’s SmartPlates are high-performance direct-to-chip liquid cooling cold plates designed to cool everything from modern processors to extreme Superchips exceeding 3,000W TDP and 500 W/cm² thermal loads. Powered by JetCool’s patented microconvective cooling® technology, SmartPlates deliver precise, high-efficiency cooling directly to chip-level hot spots—reducing thermal resistance and improving system performance.
- Supports up to 3kW TDP processors
- Outperforms traditional microchannel design by 25%, as validated by third-party testing
- Available for compute-intensive processors on the market, including AMD, Intel, NVIDIA and more
SmartSense Coolant Distribution Unit
SmartSense CDU is a high-performance, rack-mounted liquid-to-liquid Coolant Distribution Unit designed to meet the demands of GPU-dense, high-power racks. Capable of cooling up to 300kW per rack or across neighboring racks, it effectively manages heat in power-intensive environments, ensuring your data center infrastructure remains optimized for next-generation workloads.
- 300 kW of cooling per SmartSense 6U unit, with scalability up to 2.1 MW in row-level configurations
- Ultra-efficient, redundant pumps operate at low speeds (3600 RPM) for quiet, efficient, and reliable performance
- Maximize GPU utilization with warm inlet temperatures up to 60°C, run at peak performance longer, accelerate computation-intensive processes and reduce time-to-solution for AI models
Liquid Cooling for the NVIDIA Blackwell Series
Drive the next wave of AI innovation with JetCool’s direct-to-chip liquid cooling modules, designed for the NVIDIA GB200 ‘Blackwell’ Superchip. As computing evolves, so must data center cooling infrastructure. Our patented technology directly targets thermal hotspots, ensuring efficient cooling and optimized performance for AI workloads, high-performance computing, and virtualization.
JetCool’s product portfolio is designed to tackle today’s toughest chip cooling challenges, managing total power exceeding 1,500W per socket and effectively handling heat fluxes above 5W/mm².
Explore JetCool's Other Products
SmartPlate System for Dell PowerEdge Servers
SmartPlate System is a chassis-level, closed-loop, liquid-assisted air cooling solution. This system is outfitted with two enhanced cold plates that utilize our patented microconvective cooling® technology, a pump, and radiator. The cold plates directly target the processor’s hotspots, enabling the system to be more efficient and able to cool the industry’s highest power processors.
- Save an average of 15% on server power savings overnight
- Supports 850W in 1U configurations and 1,200W in 2U configurations
- Maximize compute per floor tile and minimize facility floorspace with no piping or plumbing
Direct-to-Chip Single-Phase Liquid Cold Plates for HPC
JetCool’s SmartPlates are high-performance direct-to-chip liquid cooling cold plates designed to cool everything from modern processors to extreme Superchips exceeding 3,000W TDP and 500 W/cm² thermal loads. Powered by JetCool’s patented microconvective cooling® technology, SmartPlates deliver precise, high-efficiency cooling directly to chip-level hot spots—reducing thermal resistance and improving system performance.
- Supports up to 3kW TDP processors
- Outperforms traditional microchannel design by 25%, as validated by third-party testing
- Available for compute-intensive processors on the market, including AMD, Intel, NVIDIA and more
SmartSense Coolant Distribution Unit
SmartSense CDU is a high-performance, rack-mounted liquid-to-liquid Coolant Distribution Unit designed to meet the demands of GPU-dense, high-power racks. Capable of cooling up to 300kW per rack or across neighboring racks, it effectively manages heat in power-intensive environments, ensuring your data center infrastructure remains optimized for next-generation workloads.
- 300 kW of cooling per SmartSense 6U unit, with scalability up to 2.1 MW in row-level configurations
- Ultra-efficient, redundant pumps operate at low speeds (3600 RPM) for quiet, efficient, and reliable performance
- Maximize GPU utilization with warm inlet temperatures up to 60°C, run at peak performance longer, accelerate computation-intensive processes and reduce time-to-solution for AI models
