Accelsius Announces General Availability of NeuCool IR150 and Launches NeuCool HyperStart Program

Key Takeaways

• IR150 integrates a two‑phase CDU, 42U rack, and manifolds in 800 mm enclosure
• Provides up to 150 kW cooling capacity per rack, suitable for edge AI workloads
• Two‑phase cooling can cut energy consumption by up to 90% versus air cooling
• HyperStart program gives hyperscalers engineering support to validate deployments

Pulse Analysis

Accelsius’s NeuCool IR150 marks a turning point for liquid‑cooling adoption by packaging a two‑phase coolant distribution unit, 42U of compute space, and built‑in vapor manifolds into a single 800 mm rack. This integrated design eliminates the need for separate cooling infrastructure, allowing data‑center operators—whether hyperscalers, neocloud providers, or midsize enterprises—to purchase a ready‑to‑install solution through traditional IT channels. The plug‑and‑play nature lowers entry barriers and accelerates deployment cycles for AI workloads, especially at the edge where space and power density are at a premium.

Beyond convenience, the IR150’s two‑phase technology delivers substantial sustainability benefits. By using a non‑conductive dielectric refrigerant instead of water, the system removes the risk of leaks and eliminates the massive water withdrawals associated with conventional liquid cooling. Industry studies cite up to a 90% reduction in cooling‑related energy consumption, while Jacobs Engineering analysis shows 35‑44% annual OpEx savings and 8‑17% total cost‑of‑ownership improvement over single‑phase solutions. These efficiencies help operators meet tightening community and regulatory scrutiny over water and power usage while preserving profitability.

The newly introduced NeuCool HyperStart program extends the IR150’s impact by offering structured engineering support, deployment planning, and technical validation to early adopters. By partnering with hyperscale AI cloud providers, Accelsius aims to embed two‑phase cooling into reference designs and product roadmaps, fast‑tracking large‑scale, high‑density AI deployments. This collaborative approach not only de‑riscos the technology for operators but also signals a broader industry shift toward water‑less, energy‑efficient cooling as AI compute demands continue to surge.