← Back to GPN · All articles

NVIDIA's Open-Source Wireless Tools Pass 1 Million Downloads as 6G Research Accelerates

OFFICIAL Written by GPN · Nov 7, 2026, 7:00 PM ET · NVIDIA · Tech

A year after NVIDIA open-sourced its Aerial wireless software, the company's open 5G and 6G tools have passed one million downloads, with developer adoption up tenfold over the past year. More than 8,000 research and academic institutions now use the platforms, which pair simulation and AI training with tests on real over-the-air wireless systems.

The work on show ranges from robotics to power saving. At King's College London, researchers linked a camera-equipped robot arm to an NVIDIA Jetson AGX Orin running the Sionna software, sending compact descriptions of objects instead of full images. In fruit-packing and cube-stacking tests, the approach cut task completion time by up to 52.6 percent.

SoftBank tested a neural receiver built into NVIDIA's Aerial CUDA-Accelerated RAN with a commercial 5G phone, and recorded higher uplink throughput than a conventional receiver in all 15 test settings, with gains of roughly 60 percent at best. At ETH Zurich, tuning neural receivers with local measurements cut block error rates by more than a third.

Energy is another front. A University of Leeds system predicts network demand ten minutes ahead and powers down base stations when neighbouring stations can cover the gap, cutting radio access network power use by 45.6 percent in tests against a full-power baseline.

And simulations are getting faster. The Centre Tecnològic de Telecomunicacions de Catalunya has brought GPU-accelerated radio channel models into the open-source 5G-LENA simulator, cutting a simulation that took about three hours down to 10 to 20 minutes.

NVIDIA is showing more of the work at its 6G Developer Sessions at GTC Washington, D.C.

Facts gathered from

• NVIDIA Blog

More on this

Share