6G Networks: Cisco and Nvidia Unveil AI-Native Telco Stack

Cisco has announced telecom-focused AI advances spanning radio access, core and data centre networking. Revealed at GTC in Washington DC on 28 October 2025, the portfolio centres on the Cisco N9100 switch built on Nvidia Spectrum-X Ethernet and a Nvidia Cloud Partner-compliant reference architecture for neocloud and sovereign cloud. 

For operators, Cisco, Nvidia and partners also introduced what they describe as the first AI-native wireless stack for 6G.

Jeetu Patel, President and Chief Product Officer of Cisco, says "We're at the beginning of the largest data centre build-out in history.

"The infrastructure that will power the agentic AI applications and innovation of the future requires new architectures designed to overcome today's constraints in power, computing and network performance. 

"Together, Cisco and Nvidia are leading the way in defining the technologies that will power these AI-ready data centres in all their varieties, from emerging neoclouds, to global service providers, to enterprises and beyond."

AI-native wireless stack points to 6G

As devices move beyond smartphones to AR glasses, connected cars and robotics, Cisco and Nvidia have collaborated with industry partners on an AI-RAN stack that integrates sensing and communications.

Demonstrated at Nvidia GTC DC, it lets providers infuse AI into current 5G Advanced services while laying groundwork for 6G. The stack combines Cisco’s user plane function and 5G core software with the Nvidia AI Aerial platform to enable physical AI and integrated sensing with a focus on efficiency and security.

Unified reference design for telco clouds

Cisco’s N9100 series is orderable before year-end and supports NX-OS or SONiC, allowing operators to standardise AI fabrics while retaining OS choice. The N9100 anchors Cisco’s Nvidia Cloud Partner-compliant design, managed through Cisco Nexus Dashboard across Cisco Silicon One, cloud-scale ASICs and Spectrum-X based systems.

For neocloud and sovereign cloud use, the Cisco Cloud Reference Architecture follows Nvidia Cloud Partner design tenets and adds Cisco Silicon One and cloud-scale ASIC options, plus the Cisco 8223 based on Silicon One P200 for scale-across networks. It also incorporates Nvidia BlueField-4 DPUs and Nvidia ConnectX-9 SuperNICs.

Gilad Shainer, SVP of Networking at Nvidia, says: “Nvidia Spectrum-X Ethernet delivers the performance of accelerated networking for Ethernet.

“Working with Cisco’s Cloud Reference Architectures and Nvidia Cloud Partner design principles, customers can choose to deploy Spectrum-X Ethernet using the newest Cisco N9100 series or Cisco Silicon One based switches to build open, high-performance AI networks.”

Security, observability and ecosystem for operators

For enterprise and service provider environments, Cisco is expanding the Secure AI Factory with Nvidia to prioritise performance, security and visibility. Cisco AI Defense now integrates with Nvidia NeMo Guardrails and is orderable for on-premises data-plane deployment to help limit sensitive data leaving data centres. Splunk Observability Cloud monitors AI stack health with Cisco AI PODs and Splunk Enterprise Security extends protection to AI workloads.

Core infrastructure additions include Cisco Isovalent validation for inference on AI PODs and Cisco Nexus Hyperfabric AI with a cloud-managed Cisco G200 Silicon One switch delivering high-density 800G Ethernet. 

Cisco UCS 880A M8 with Nvidia HGX B300 and Cisco UCS X-Series with Nvidia RTX PRO 6000 Blackwell Server Edition GPUs are now orderable to support fine-tuning and inference. Cisco’s ecosystem now includes Nvidia Run:ai for GPU orchestration, plus supported Nutanix Kubernetes Platform, Nutanix Unified Storage and Nutanix Enterprise AI. Cisco is also aligning with the Nvidia AI Factory for Government for regulated deployments.

Partner views from the field

Industry voices underline operational simplicity as AI scales on telco-grade networks.

Xiaohe Hu, CEO of  Infrawaves, says: The real challenge in AI infrastructure isn't just performance — it's maintaining operational sanity as you scale from dozens to thousands of GPUs. 

Cisco's approach with NX-OS and Nexus Dashboard creates a single pane of glass across our entire AI fabric, whether we're optimising inference latency in the front-end or maximizing training throughput in the back-end. That operational simplicity translates directly to faster deployments and lower TCO."