Designed to run across AI clouds and modern datacenter infrastructure, on any GPU and emerging XPUs, NR-NEXUS launches with beta customers ahead of full commercial availability later this year

NeuReality, a pioneer in AI infrastructure, today introduced NR-NEXUS, an inference operating system designed to power large-scale inference services. Already deployed with beta customers, NR-NEXUS enables organizations to transform fragmented systems into production-ready token factories.

The platform was developed based on NeuReality’s deep expertise in AI hardware architecture and large-scale inference system design. It marks the next step in building the foundation for modern AI inference at scale.

NR-NEXUS is a hardware-agnostic operating system for AI Factories that works across any CPU, GPU, or NIC, and supports enterprise-scale AI deployment. Just as the PC was the computer of the internet era, the AI factory is the new computer, the core infrastructure unit powering the intelligence era.

The growing demand for inference fluctuates constantly, often leaving GPUs underutilized and infrastructure fragmented across multiple runtimes and systems. These inefficiencies increase costs, reduce performance, and limit the return on AI infrastructure investments. NR-NEXUS addresses this by allowing organizations to run inference across hyperscale cloud environments, dedicated GPU clusters, and emerging XPUs, all without re-architecture or disruption to existing deployments.

By orchestrating the full inference stack through a unified platform, NR-NEXUS increases utilization, stabilizes performance, and lowers the cost of generating tokens.

“AI inference is rapidly becoming one of the largest computing markets in the world, yet the infrastructure stack around it remains fragmented,” said Moshe Tanach, CEO of NeuReality. “With NR-NEXUS, we are defining the operating system for AI token factories – enabling organizations to run and scale inference workloads efficiently across GPUs, emerging XPUs, hyperscalers, and dedicated AI clusters. As open-source models and AI-native applications proliferate, operators need infrastructure that gives them flexibility rather than lock-in. NR-NEXUS provides that foundation.”

NR-NEXUS is designed for NeoCloud providers, enterprises, and semiconductor vendors looking to consolidate siloed infrastructure into complete inference platforms accelerating time to market with new AI models and maximizing ROI of AI factory builds. Learn more about NR-NEXUS at www.neureality.ai/nexus or meet the NeuReality team at NVIDIA GTC.

About NeuReality

Founded in 2019, NeuReality is a pioneer in purpose-built inference infrastructure for AI factories. Based on an open, standards-based approach, NR-NEXUS^®, NR2^® AI-SuperNIC, NR1^® AI-CPU and NR1^® Inference Appliance are fully compatible with any hardware. It employs 80 people across facilities in Israel, Poland, and the U.S. To learn more, visit http://www.neureality.ai.

View source version on businesswire.com: https://www.businesswire.com/news/home/20260312503240/en/

"With NR-NEXUS, we are defining the operating system for AI token factories – enabling organizations to run and scale inference workloads efficiently across GPUs, emerging XPUs, hyperscalers, and dedicated AI clusters." - Moshe Tanach, CEO of NeuReality