1Finity Ships Full-Stack AI-Native vRAN Built on Shared NVIDIA GPU Infrastructure

1Finity, a Fujitsu company, announced on November 3, 2026 the availability of a full-stack AI-Native vRAN solution that lets mobile network operators run AI and RAN workloads on shared NVIDIA GPU infrastructure. The company said the integrated stack is aimed at operators seeking to sell AI-driven enterprise services alongside private wireless networks. The announcement was distributed by Business Wire.

What 1Finity's AI-Native vRAN Stack Includes

1Finity said the ready-to-deploy stack combines its own Open vRAN software and O-RAN compliant radio units with high-performance AI servers from Quanta Cloud Technology and Supermicro that support the scalable NVIDIA AI Aerial platform for integrated AI and RAN workloads. The stack also includes a no-code platform from Aible that lets business users build, deploy and operate AI agents using enterprise data, with guided workflows and more than 100 customizable templates, and Bridge by Armada, which provides multi-tenant orchestration, security and billing for large-scale compute deployments.

1Finity said the shared GPU architecture lets operators deploy AI capabilities more economically than building separate, siloed compute environments, with improved resource utilization, lower costs and faster time to market. The company said operators have not been able to fully realize over-the-top revenue from AI-driven enterprise applications, and that the new stack lets them deliver private wireless networks alongside AI-driven applications.

"By enabling operators to deliver complete AI solutions — from infrastructure to applications — 1Finity is redefining how telecom networks create and capture value in the AI era," said Patrik Eriksson, vice president and head of the mobile systems business unit at 1Finity.

Soma Velayutham, vice president of AI and telecoms at NVIDIA, said: "By building its full-stack enterprise solution on the NVIDIA AI Aerial platform, 1Finity is empowering MNOs with high-performance, software-defined AI-native infrastructure that seamlessly processes AI and RAN workloads."

1Finity's Earlier AI-RAN Work With NVIDIA

According to akta.pro company records, 1Finity delivered a live AI-RAN field-validated blueprint with Amdocs, Supermicro, NVIDIA and Red Hat in June 2026, integrating 1Finity Open vRAN with Amdocs aOS agentic OS on NVIDIA GH200 Grace Hopper Superchips for AI-assisted RAN optimization. The same records list a June 2026 partnership in which Supermicro provides single-rack-unit servers configured for AI-RAN workloads using NVIDIA GH200 Grace Hopper processors, running 1Finity Open vRAN software on Red Hat OpenShift.

How 1Finity Built Toward the Launch

In October 2026, 1Finity argued that operators should sell idle AI-RAN compute to enterprises first, citing underutilized capacity and a business case, and proposed policy-driven orchestration to reserve RAN capacity and then partition and bill the remaining resources, Telecompaper reported. In August 2026, 1Finity announced integration of its IP over DWDM infrastructure into the New Mexico Fiber Network, a statewide middle-mile backbone built from assets of nine local exchange carriers, according to Yahoo. In July 2026, Chunghwa Telecom, Fujitsu Limited and 1Finity announced a two-year strategic partnership to expand All-Photonics Network deployment and explore quantum technologies in Taiwan, Gurufocus reported.

NVIDIA's November 17 Results Date

NVIDIA's third-quarter fiscal 2027 results are scheduled for November 17, 247wallst reported.