LITEON and NVIDIA Partner on AI-RAN Tech
At MWC Barcelona 2026, LITEON Technology is showcasing its latest advancements in AI-RAN (Radio Access Network). The company is integrating its Open Radio Units with NVIDIA's AI Aerial platform to accelerate the commercialization of AI-native, GPU-accelerated network architectures. This collaboration aims to help the telecom industry transition to more scalable and efficient network solutions.
The move toward AI-RAN represents a fundamental shift from traditional, hardware-centric Radio Access Networks to more flexible, software-defined infrastructures. This allows for running both cellular and AI workloads on the same general-purpose hardware, a convergence that promises greater efficiency and cost-effectiveness. The AI-RAN market is projected to grow significantly, with one report estimating it will reach $816.04 million by 2032, up from $150 million in 2024. At the core of this partnership is the NVIDIA Aerial™ platform, a suite of software and tools designed to build and deploy AI-native wireless networks. It enables the dynamic allocation of 5G and AI tasks on the same GPU, which can increase capacity utilization by two to three times. This software-defined approach is considered a crucial step for preparing networks for the evolution to 6G. LITEON's contribution includes its O-RAN compliant 5G open-radio units (O-RUs). These units are designed with an open architecture for rapid integration and are focused on high performance and energy efficiency. The collaboration ensures LITEON's hardware is fully compatible with NVIDIA's AI Aerial platform, aiming to reduce integration complexity and speed up deployment for telecom operators. This collaboration is part of a broader industry trend, with an entity known as the AI-RAN Alliance, which was established in 2024, working to accelerate the integration of AI into radio networks. Key players in the AI-RAN market include traditional telecom giants like Ericsson and Nokia, as well as technology companies specializing in AI. The push towards Open RAN initiatives is a significant factor, promoting interoperability between hardware and software from different vendors. The real-world impact of GPU-acceleration in 5G networks has already been demonstrated. In a deployment with NTT DOCOMO, a similar GPU-based solution resulted in a reduction of total cost of ownership by up to 30% and lowered power consumption at base stations by up to 50%. For end-users, this technology enables ultra-low latency applications like autonomous vehicles, remote surgery, and real-time AI processing at the network edge. LITEON is also working with partners like SynaXG and Supermicro to integrate RAN and AI workloads under a unified architecture. This ecosystem collaboration is seen as essential for moving AI-RAN from trials to commercial viability. Richard Chiang, a General Manager at LITEON, emphasized that this open architecture allows operators to deploy AI capabilities directly within their 5G networks.
