Nvidia, Rebellions and SK Telecom said Nov. 9 they will collaborate to advance Korea’s sovereign AI infrastructure, using Nvidia’s NVLink Fusion to connect Korean-designed processors to Nvidia systems, Nvidia announced. The Korea Herald reported, citing the Ministry of Science and ICT, that the companies signed a memorandum of understanding on Monday at Nvidia’s AI Day Seoul event at Coex.
Why it matters: Nvidia said Rebellions plans to use NVLink Fusion to connect its next-generation Rebel R200 neural processing units to Nvidia’s rack-scale inference platform and to use the Nvidia MGX rack architecture. SK Telecom plans to deploy those systems in its AI data center, Nvidia said, as part of a full-stack sovereign AI effort that includes infrastructure, proprietary models and public-facing services. The Korea Herald said the ministry described the deal as a way to bring locally developed AI processors into Nvidia data-center systems and to help data centers balance performance and power use by combining NPUs and GPUs for inference.
Nvidia CEO Jensen Huang said, “With NVLink Fusion, Korea’s chipmakers can connect their custom ASICs to NVIDIA’s AI factory platform, tightly coupling their processors with NVIDIA compute to run AI workloads.” Rebellions cofounder and CEO Sunghyun Park called the announcement “a turning point” for connecting Korean AI silicon and models with global systems technology. The Herald said the ministry wants to extend a similar model to other domestic NPU makers and has also been pursuing links with AMD.