Announcing at NVIDIA GTC 2026, the IR150 delivers 150kW of cooling capacity in a single, pre-integrated enclosure, reimagining infrastructure for the liquid cooling era.
SAN JOSE, Calif.--(BUSINESS WIRE)--Accelsius, the leader in two-phase, direct-to-chip liquid cooling technology for AI and high-performance computing, today introduced the NeuCool® IR150, the industry’s first fully integrated rack-level cooling solution that combines a two-phase Coolant Distribution Unit (CDU), 42U of IT rack space, and built-in liquid and vapor manifolds in a single 800mm-wide enclosure, offering up to 150kW of capacity.
“The IR150 represents the next evolution of data center infrastructure,” said Josh Claman, CEO of Accelsius.
The announcement arrives at a pivotal moment for the data center industry. According to various sources, the global liquid cooling market is projected to be around $6 billion in 2026, driven by AI workloads that generate rack densities that air cooling can no longer manage. At the same time, NVIDIA CEO Jensen Huang revealed earlier in 2026 that the company’s next-generation Vera Rubin platform, expected in the second half of 2026, is designed to be cooled entirely via liquid cooling, using warm overhead facility water at temperatures up to 45°C. As the industry’s leading chipmaker signals a future built on liquid cooling, Accelsius, an NVIDIA Inception program member, is delivering the infrastructure to make that future a reality today.
Reimagined for the mission-critical liquid cooling era, the IR150 integrates a two-phase CDU directly into the IT enclosure, allocating 200mm to cooling infrastructure and 600mm to server space. The result is a true plug-and-play solution for high-density, two-phase-enabled deployments. The system dramatically simplifies installation, reduces technology cooling system (TCS) complexity and cost, and confines failure domains to a single rack.
“The IR150 represents the next evolution of data center infrastructure,” said Josh Claman, CEO of Accelsius. “When the world’s leading chipmaker designs its next-generation AI supercomputers to run on liquid cooling, the industry needs purpose-built infrastructure to deliver it. The IR150 does exactly that, a single, integrated rack that arrives ready to cool the most demanding AI workloads, with minimal chiller infrastructure, no water treatment, and no compromise on reliability.”
Traditional single-phase liquid cooling systems rely on treated water circulated directly to the chip, introducing leak risk, corrosion concerns, and continuous water-quality maintenance.
In contrast, Accelsius’ two-phase approach uses a non-conductive dielectric refrigerant with an A1 safety rating and low global warming potential. No water enters the IT rack, meaning leak events pose minimal risk to GPUs or server electronics. Industry studies have shown that two-phase cooling systems can reduce cooling energy consumption by up to 90 percent and eliminate millions of gallons of annual water use compared to air-cooled alternatives, while independent analysis by Jacobs Engineering has demonstrated that Accelsius’ two-phase solutions deliver 35–44 percent annual OpEx savings and 8–17 percent five-year total cost of ownership savings over single-phase direct-to-chip systems.
Key benefits of the NeuCool IR150 include:
- Pre-integrated, plug-and-play design: CDU, 42U IT rack space, and liquid and manifolds ship as a single, factory-tested unit, reducing deployment time and on-site complexity.
- 150kW of two-phase cooling capacity: Supports current-generation GPUs including NVIDIA Blackwell and is designed with thermal headroom for next-generation platforms such as NVIDIA Vera Rubin.
- Single-rack failure domain: Modular architecture isolates cooling to each rack, simplifying maintenance and improving uptime for mission-critical environments.
- Warm-water compatible: ASHRAE W45 compatible, supporting facility water temperatures up to 45°C to enable more free-cooling days and reduce energy use.
- Zero water in the rack: Dielectric refrigerant minimizes leak risk to IT equipment, removes water treatment requirements, and supports sustainability goals.
- Hot-swappable serviceability: Pumps, power supply units, sensors, main control board, and LCD display can all be serviced without taking the rack offline.
See the NeuCool IR150 at NVIDIA GTC 2026
Learn more about the NeuCool IR150 at NVIDIA GTC 2026, March 16–19 Booth 129 at the San Jose McEnery Convention Center. To schedule a meeting or demo, contact info@accelsius.com.
The IR150 joins the NeuCool MR250, Accelsius’ row-based CDU delivering up 250kW+ of cooling capacity per rack, and the NeuCool Thermal Simulation Rack (TSR), a first-of-its-kind thermal test platform. Together, the NeuCool product family offers operators a complete suite of two-phase, direct-to-chip cooling solutions that scale from evaluation through full data center deployment.
About Accelsius
Founded by Innventure, Inc. (NASDAQ:INV), Accelsius empowers data center and AI neoclouds to achieve their business, financial and sustainability goals through advanced cooling solutions. The proprietary NeuCool® platform provides best-in-class thermal efficiencies through a safe, two-phase, direct-to-chip liquid cooling system that scales from single racks to entire data centers. For more information, visit accelsius.com or follow us on LinkedIn.
