SynaXG Proves the Art of the Possible with Fully Software-Defined AI-RAN on NVIDIA AI Aerial
BARCELONA, Spain and SINGAPORE, March 02, 2026 (GLOBE NEWSWIRE) — SynaXG immediately introduced main technical breakthroughs in AI-native radio entry networks, demonstrating concurrent operation of 5G FR1, 5G FR2 and AI workloads on NVIDIA AI-RAN platform with real-time, policy-driven GPU orchestration. The outcomes affirm that software-defined, AI-native RAN can ship carrier-grade efficiency and steady 24×7 operation, prepared for industrial deployment.
SynaXG achieved carrier-grade 5G FR1 efficiency on a single NVIDIA GH200 platform. The system concurrently operated 20 x 100MHz 5G NR cells, delivering greater than 36Gbps of aggregated throughput with sub-10 millisecond latency, whereas supporting as much as 1,200 linked consumer tools per cell. These outcomes validate {that a} absolutely software-defined AI-RAN working on accelerated computing infrastructure can match the efficiency of best-in-class industrial 5G deployments, whereas enabling AI workloads to run concurrently on the identical shared platform with out compromising community KPIs.
SynaXG additionally achieved the world’s first carrier-grade FR2 virtualized RAN implementation working concurrently with FR1 and AI workloads on a shared GPU platform. The system delivered excessive throughput with end-to-end latency as little as 5 milliseconds, assembly the efficiency necessities for dense city and enterprise deployments. This milestone demonstrates that millimetre-wave RAN will be realized as a completely software-defined workload on accelerated infrastructure with out compromising carrier-grade KPIs.
SynaXG’s real-time orchestration software program dynamically allocates, and switches Multi-Occasion GPU (MIG) partitions based mostly on stay RAN and AI key efficiency indicators. GPU sources are repeatedly rebalanced throughout FR1 for protection, FR2 for capability and AI workloads with out service interruption, enabling deterministic RAN efficiency whereas maximizing infrastructure utilization.
The AI-RAN system operated repeatedly, 24 hours a day, 7 days every week, below sustained load, validating the steadiness, scalability and operational resilience required for service environments. The outcomes affirm that NVIDIA GPU-accelerated AI-RAN can preserve predictable efficiency whereas adapting dynamically to altering visitors situations and workload demand.
SynaXG’s AI-RAN platform is constructed on {hardware} and software program stack leveraging NVIDIA AI Aerial platforms, enabling a software-defined, AI-native structure that helps deterministic community efficiency alongside AI processing on shared GPU infrastructure.
Wanting forward, SynaXG is extending its AI-RAN platform to extra NVIDIA platforms, together with NVIDIA DGX Spark, demonstrating the portability of its software-defined structure. This write as soon as, run anyplace method throughout NVIDIA CUDA-based platforms accelerates innovation cycles and simplifies deployment at velocity that has by no means been seen with conventional approaches.
In parallel, SynaXG continues to advance AI-for-RAN capabilities centered on community optimization and automation, spectral and power effectivity, ISAC, distributed MIMO, ICIC, and many others. SynaXG has been working with companions on AI brokers that analyse real-time community situations and repair demand and take actions to boost efficiency, power effectivity and operational simplicity in AI-native networks.
“With the current industry-leading breakthroughs, SynaXG has demonstrated that AI-RAN can ship carrier-grade FR1 and FR2 efficiency with steady 24×7 operation on shared NVIDIA AI infrastructure,” mentioned Xin Huang, CEO of SynaXG. “Our software-defined method permits operators and enterprises to deploy scalable, AI-native networks and are prepared for industrial deployment.”
“Software program-defined structure is vital to the following technology of wi-fi networks. SynaXG’s benchmark 5G efficiency on NVIDIA AI-RAN platform proves that operators and enterprises can obtain the pliability and agility of cloud-native computing whereas sustaining carrier-grade throughput and efficiency per watt, required for industrial 5G companies”, mentioned Soma Velayutham, VP of AI and Telecoms, NVIDIA.
SynaXG will showcase these AI-RAN breakthroughs as a part of the AI-RAN Alliance Working Group 2 demonstrations at AI-RAN Alliance sales space Corridor 2, 2E45 and on the SynaXG sales space at Corridor 5, 5F21 Cell World Congress 2026 in Barcelona.
About SynaXG
SynaXG is a Singapore-based expertise firm pioneering AI-native radio entry networks (AI-RAN). Its unified software program platform integrates 5G RAN and AI capabilities on shared industrial {hardware}, enabling operators, hyperscalers and enterprises to deploy high-performance, energy-efficient, and monetizable AI-driven networks. SynaXG works intently with world ecosystem companions throughout silicon, cloud, RAN and AI to speed up the transition towards 5G-Superior and 6G architectures.
For extra info, go to http://www.synaxg.com
Watch the video accompanying this announcement on YouTube.
For added context, see NVIDIA’s AI-RAN on its company weblog.
Media contact: media@synaxg.com
A video accompanying this announcement is on the market at https://www.globenewswire.com/NewsRoom/AttachmentNg/e0e6c93a-fb23-4f53-9edb-f495916f1c41
