NVIDIA says cities and partners are building AI services on its technology

NVIDIA said Nov. 3 that partners in Spain, Poland, Malaysia and elsewhere are building city services on its AI models and GPUs, in a post published for Barcelona's Smart City Expo World Congress1.

NVIDIA's argument is about cost of entry and data control. The post says shared computing infrastructure lets city departments, neighboring municipalities and local developers extend services without each acquiring and managing its own resources. It says local governments stay in control of how sensitive data is processed, and that some city teams can start on NVIDIA RTX workstations or a small GPU system, depending on the workload.

The post builds on initiatives announced at GTC Berlin. The Expo runs Nov. 3-5, and NVIDIA said its Vice President of EMEA, Paolo Guglielmini, will deliver a keynote there.

The examples differ in maturity, and not all are municipal services.

Deployed, per the companies. NVIDIA said 1MillionBot combines open-source ALIA models with NVIDIA Nemotron models in resident-service agents for Madrid Digital, Administració Oberta de Catalunya and 150+ deployments across Spain. LiveX says5 its AI City Hosts are deployed at over 100 customer sites, including PUMA stores, San José Mineta International Airport and El Camino Hospital. Those details come from LiveX's own Nov. 3 release, not NVIDIA's post.

Built or planned. NVIDIA said K2K is building shared AI infrastructure in Wrocław, Poland, for video search and summarization across several city functions. ITMAX, which sells managed services to Malaysian cities, is scaling its SamurAI V2 vision agents on NVIDIA RTX PRO 6000 GPUs to support 17,000+ cameras by 2027, a target rather than current scale. At Dangote's Lekki refinery in Nigeria, Honeywell Technologies' Forge Plant Cognition uses Nemotron-powered agents, and a digital twin of the plant is planned. Cires Technologies is expanding a sovereign AI factory at Morocco's Tanger Med port for planned worker-safety, congestion, cargo-routing and port applications.

NVIDIA's post also pointed to the Barcelona Supercomputing Center's AI Factory Accelerator Programme, which gives startups access to sovereign computing, technical support and training. The post does not name an NVIDIA product in connection with it.

Performance claims. The post says ITMAX's use of the TeraSim traffic simulator to test signal timing cut peak-hour corridor travel times by up to 26%. It also says Linker Vision and MVRDV's Omniverse and Nemotron agents cut design iteration from two weeks to under an hour. These are company claims, and the post includes no independent testing.

Access terms. The post does not include pricing. NVIDIA's Nemotron page says the models can be downloaded from Hugging Face for free2, while NIM microservice deployment requires an NVIDIA AI Enterprise license. NVIDIA's Omniverse page says4 that, as of May 2026, Omniverse is free for development and production use without an AI Enterprise subscription.