Nvidia Says Use of Its Open 6G Research Software Has Grown Tenfold in a Year

More than 8,000 research and academic institutions now use the platforms, the company said. The figures matter for Nvidia's effort to position its GPUs as the base for AI-native mobile networks, in which radio processing, computing and sensing run on shared, software-defined hardware rather than on dedicated telecom equipment.

The post set out results that research groups have reported using the tools. At King's College London, researchers connected a camera-equipped robot arm to an Nvidia Jetson AGX Orin running Sionna and had it send compact descriptions of objects and their positions rather than full images. In fruit-packing and cube-stacking experiments with deliberately introduced faults, the approach cut task completion time by up to 52.6% and improved success rates compared with sending periodic images, according to Nvidia.

SoftBank integrated a neural receiver into Nvidia's Aerial CUDA-Accelerated RAN and tested it with a commercial 5G smartphone. Under controlled interference, it delivered higher uplink throughput than a conventional receiver at all 15 settings tested, with a maximum gain of roughly 60%, Nvidia said. Researchers at ETH Zurich found that tuning neural receivers with local measurements cut block error rates by more than a third compared with selected pretrained versions.

At the University of Leeds, a system that predicts network demand 10 minutes ahead and uses those forecasts to put base stations to sleep reduced radio access network power consumption by 45.6% against a full-power baseline while keeping devices connected. The team tested it on a portable 5G network-in-a-box powered by Nvidia's DGX Spark.

Spain's Centre Tecnològic de Telecomunicacions de Catalunya added GPU-accelerated radio channel models to its open source 5G-LENA simulator, cutting simulation runtimes from about three hours to between 10 and 20 minutes in an early evaluation.

The results are reported by Nvidia and its research partners and have not been independently replicated. Standards work on 6G is still under way, with commercial networks generally expected around the end of the decade.