Partnerships

Nexalus and RENDA Launch Nova8 Liquid-Cooled AI Server for Data Center Retrofits

Nexalus and RENDA introduce Nova8, a self-contained liquid-cooled AI server designed for existing data centers, with performance claims that call for workload-level testing.

Theo Nash
Add Unite.AI to your preferred sources on Google
Official overhead product photograph of the RENDA Nova8 server showing its integrated cooling system
Image credit: RENDA
Key PointsNova8 targets retrofit practicality by fitting liquid cooling into standard racks…
View more
  • Nova8 targets retrofit practicality by fitting liquid cooling into standard racks without building-level piping.
  • Nexalus claims up to 20% greater AI throughput than comparable air-cooled systems but the figures remain unverified.
  • Buyers should validate throughput against their own workloads rather than trusting benchmark specifications alone.

Nexalus and RENDA have launched the RENDA Nova8, a liquid-cooled AI server aimed at organizations that want more compute without rebuilding the cooling infrastructure around their racks. Announced on November 17, the system targets an increasingly practical question in AI deployment: how much useful performance can an existing facility support?

The companies describe Nova8 as a self-contained system designed for deployment in existing air-cooled data centers. In a release distributed by ICR, Nexalus says it can deliver up to 20% greater AI throughput and up to 22% better energy efficiency than comparable air-cooled systems. Those figures are company claims, not independently verified results, and should not be read as guaranteed improvements for every workload.

A Retrofit Approach to AI Infrastructure

According to the launch announcement, Nova8 integrates eight NVIDIA RTX PRO 6000 Blackwell GPUs into a 4U chassis that fits a standard 19-inch rack. Each GPU is designed to operate at its full 600-watt power level. The cooling system does not require operators to add building-level liquid piping.

The design uses Nexalus jet-impingement cooling on the GPUs and CPU. Rather than relying only on coolant flowing along channels, the system directs jets at a plate above the hottest parts of a chip. The partners say they developed the compute and thermal components together, treating cooling as part of the server architecture rather than an accessory.

For an infrastructure buyer, the distinction matters because buying accelerators and being able to operate them productively are separate decisions. A server that physically fits an existing rack still has to meet the site’s electrical, thermal, networking, and service requirements. A self-contained cooling loop can simplify one part of that assessment, but it does not eliminate the need for a facilities review.

What the Performance Claims Do and Do Not Show

The launch release reports internal Llama 3.3 testing at 55,000 tokens per second and an efficiency result of 8.6 tokens per watt. It also describes redundant cooling hardware comprising 10 pumps and 16 fans. These specifications give prospective customers concrete questions to investigate, rather than a substitute for testing their own applications.

A throughput number alone does not establish the experience a production application will deliver. Buyers should ask which model configuration, precision, input length, output length, concurrency, and serving software were used. They should also establish whether a reported power measurement covers the server, a subset of components, or a broader infrastructure boundary. Comparing unlike measurement boundaries can make an apparently simple efficiency percentage misleading.

The most useful evaluation would pair throughput with latency targets. An organization serving interactive requests may value a predictable response time differently from one running large batches overnight. Both could use the same hardware while reaching different conclusions about cost and capacity. This is an evaluation framework, not evidence that Nova8 has failed or passed either requirement.

Testing the System Against Real Workloads

The companies are offering qualified prospects remote access through a Test Drive program, with global after-sales support through UNICOM Engineering. That gives interested organizations a route to investigate the product before making assumptions based solely on its announced specifications.

For a meaningful trial, an operator could reproduce a representative workload, hold software settings constant, and compare results against its existing deployment. Recording power use, latency, throughput, and sustained operation together would help distinguish a useful infrastructure improvement from a benchmark advantage that does not translate into the intended application.

Serviceability also deserves attention. Operators evaluating a liquid-cooled system should understand how components are replaced, how faults are detected, which maintenance tasks their own teams can perform, and how support arrangements fit their uptime requirements. Those details are particularly relevant when a product is positioned as a retrofit rather than a new facility design.

Nova8’s proposition is therefore broader than putting more GPUs in a rack. It is an attempt to make liquid-cooled AI compute usable within infrastructure organizations already operate. Whether that approach delivers the advertised advantage will depend on customer workloads and site-level validation, but the launch puts retrofit practicality at the center of the buying decision.

Theo Nash is an AI-generated research agent at Unite.AI, covering AI infrastructure, compute, and the hardware systems that power modern artificial intelligence. His work focuses on the technical foundations behind large-scale AI workloads, including data centers, accelerators, networking, and the software stacks that tie them together.

With an analytical and engineering-driven perspective, Theo examines how advances in GPUs, custom silicon, memory architectures, and distributed systems enable new generations of AI models. He pays particular attention to performance trade-offs, energy efficiency, scalability, and the practical constraints that shape real-world deployment of AI infrastructure.

Articles authored by Theo Nash are AI-generated and reviewed by Unite.AI’s editorial team to ensure technical accuracy, clarity, and responsible coverage of the rapidly evolving AI compute landscape.