Naver Cloud, Nvidia Expand Alliance to Build Global AI Factories

Naver Cloud is teaming up with Nvidia to push into the global AI factory market. The companies plan to deepen cooperation across AI infrastructure, models, services and physical AI as they seek to expand their influence in the global race to build AI infrastructure.

Naver Cloud Chief Executive Officer Kim Yu-won outlined the partnership at Nvidia's Cloud Partner Summit in Taiwan on June 2. "Naver Cloud has strong full-stack technology capabilities spanning everything from AI infrastructure to services, making it a partner that fully aligns with Nvidia's AI factory platform strategy covering energy, chips, infrastructure, models and applications," Kim said.

Nvidia Chief Executive Officer Jensen Huang formally named Naver Cloud a key partner in the global AI ecosystem during his keynote at GTC Taipei 2026 in Taiwan on June 1.

The partnership centers on cooperation across the full AI stack. Naver Cloud will use Nvidia's open large language model Nemotron-3 Ultra to advance its own hyperscale AI model, HyperCLOVA X. The companies also plan joint research on large language model optimization and underlying technologies.

The two companies are also collaborating in physical AI. In March, Naver Cloud unveiled its Seoul World Model using Nvidia's physical AI platform Cosmos to recreate real-world data from Seoul. Built on domestic map data, the model was trained on 1.2 million panoramic images collected across the city and is designed to replicate South Korea's actual road environment and spatial structure.

Naver Cloud is positioning its in-house full-stack technology and dedicated cloud capabilities as key differentiators. Its strategy is to provide AI environments tailored to each customer's workload and requirements, while supporting the development of sovereign AI models for different regions through partnerships with governments and local companies.

Sovereign AI refers to efforts by countries to build their own AI models and services suited to local data, language and regulatory conditions. Naver Cloud plans to target that demand by combining its experience in infrastructure, models and service operations.

The companies plan to announce more detailed cooperation measures soon, including execution plans for the global AI factory business, following a meeting in South Korea between Naver Chairman Lee Hae-jin and Huang.

Kim said the AI industry is shifting from a model-centered paradigm to an inference-centered AI factory era that depends on the stable operation of large-scale infrastructure. As infrastructure, model and service companies expand across business lines and integration accelerates, Naver Cloud can respond fastest to the rapidly changing competitive landscape because it has directly operated every layer of the AI ecosystem, he added.

He said the partnership with Nvidia goes beyond a simple relationship between a GPU supplier and a customer. It is a strategic decision to jointly develop AI technologies and expand the global AI ecosystem. Naver Cloud aims to become a core supplier supporting surging AI demand in Asia and establish itself as a differentiated AI infrastructure hub.

Raj Mirpuri, Nvidia's vice president for global AI cloud and infrastructure, said AI factories require an ecosystem in which accelerated computing, models, data, applications and cloud services are tightly connected. Through the partnership with Naver Cloud, Nvidia will help customers in Asia and around the world make broader use of its integrated AI platform to build sovereign AI, industrial AI and enterprise AI.

Kim Dae-young, Hankyung.com reporter kdy@hankyung.com