NVIDIA, Rebellions and SK Telecom Announce Collaboration to Advance Sovereign AI Infrastructure With NVIDIA NVLink Fusion

NVIDIA, Rebellions and SK Telecom have announced a collaboration to strengthen South Korea’s sovereign AI infrastructure by combining NVIDIA’s open infrastructure platform, Rebellions’ locally developed AI processors and SK Telecom’s capabilities in AI data centers, models and services. The collaboration will leverage NVIDIA NVLink Fusion™ to integrate these technologies.

Under the agreement, Rebellions will use NVLink Fusion to connect its Rebel R200 neural processing units (NPUs) with the NVIDIA platform and adopt the NVIDIA MGX™ rack architecture to develop rack-scale systems. SK Telecom will support deployment by leveraging its expertise in AI data center operations, along with its proprietary AI models and services.

With support from South Korea’s Ministry of Science and ICT (MSIT), the initiative aims to develop AI services aligned with the country’s language, industrial needs and strategic priorities. It will bring together domestic semiconductor development, advanced computing infrastructure and operational expertise to support AI deployment at scale.

“Korea has the talent, technology and industrial ambition to become a world leader in AI,” said Jensen Huang, founder and CEO of NVIDIA. “Sovereign AI gives nations the power to turn their language, culture and expertise into intelligence. 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. Together with MSIT, Rebellions and SK Telecom, we’re building the computing foundation for Korea to produce its own intelligence and power its next industrial revolution.”

“Today’s announcement marks a turning point, opening a new chapter for sovereign AI by connecting Korea’s AI silicon and models with global systems technology,” said Sunghyun Park, cofounder and CEO of Rebellions. “By integrating Rebellions’ purpose-built silicon and system design with NVIDIA’s NVLink Fusion, high-speed interconnects and rack-scale design expertise, and SKT’s model and data center operations capabilities, we will deliver low-latency and energy-efficient inference for frontier models at scale. As a leading Korean AI silicon startup, we will lead the charge to establish Korean technology as a core pillar powering AI worldwide.”

“This collaboration is meaningful in that it validates the potential of next-generation sovereign AI infrastructure by combining global AI infrastructure technology with Korean AI semiconductors,” said Jung Jai-hun, president and CEO of SK Telecom. “By linking SKT’s AI data centers, proprietary AI models including A.X K2, and public-facing AI services developed under the ‘AI for All’ project, we will strengthen our full-stack sovereign AI competitiveness and contribute to the growth of Korea’s AI ecosystem.”

Scaling Silicon Innovation With NVIDIA’s Open Infrastructure Platform

NVLink Fusion expands NVIDIA’s vertically integrated yet horizontally open AI platform to support custom XPUs. The technology allows chipmakers to integrate specialized processors with NVIDIA’s high-bandwidth, low-latency scale-up networking and established infrastructure ecosystem. This approach is designed to enhance performance, minimize deployment risks and shorten time to market.

Rebellions is developing its next-generation R200 neural processing unit (NPU), designed to deliver low-latency, energy-efficient inference for advanced AI models, including multitrillion-parameter mixture-of-experts models. Using NVLink Fusion, the company will connect multiple R200 NPUs within a rack through NVIDIA NVLink™ scale-up networking.

The NVIDIA MGX rack architecture and reference designs will provide the foundation for system development. In addition, the MGX ecosystem will give Rebellions access to an established supply chain of original design manufacturers (ODMs) and component suppliers.

Connecting Infrastructure, Models and Services

SK Telecom contributes its expertise in developing and operating AI data centers, supported by its proprietary foundation models and consumer-facing AI services. Building on its long-standing partnership with NVIDIA, the company plans to integrate these capabilities and deploy the systems in its AI data center using the standardized MGX rack architecture. The initiative aims to deliver a full-stack sovereign AI infrastructure.

The MGX rack architecture allows GPU- and NPU-based systems to share common data center design elements, including rack dimensions, power delivery and cooling systems. This standardized approach is expected to support infrastructure planning while Rebellions continues developing its processors and systems.

The companies also plan to develop a disaggregated inference approach, which assigns different stages of AI response generation to the processor architecture best suited for each task. The proposed configuration will use NVIDIA GPUs for prefill and Rebellions NPUs for decode, with the aim of increasing inference throughput and improving power efficiency.

SK Telecom intends to integrate the infrastructure with its proprietary foundation models and consumer-facing AI services, including initiatives under the government's AI for All project, further advancing its full-stack sovereign AI strategy.

Supporting Sovereign AI Infrastructure With NVLink Fusion

South Korea’s Ministry of Science and ICT (MSIT) is supporting partnerships between NVIDIA and Korean NPU companies, starting with Rebellions and SK Telecom, to further develop the country’s sovereign AI infrastructure.

“This collaboration with NVIDIA marks a significant step forward, as it enables Korean-developed NPUs to enter the computing infrastructure market alongside NVIDIA, a global leader in AI computing, through NVLink Fusion,” said Bae Kyung-hoon, Korea Deputy Prime Minister, and Minister of Science and Information and Communication Technologies. “The government will actively support this collaboration to deliver tangible results and further advance cooperation with NVIDIA across a range of areas, including physical AI and securing GPU supply.”

The collaboration aims to build a foundation for sovereign AI services across South Korea’s public and private sectors by integrating locally developed processors and AI models with NVIDIA’s infrastructure and SK Telecom’s data center expertise.

Availability

Rebellions’ R200 systems are expected to be ready for deployment in 2028.