Rebellions Secures $400M, Launches AI Racks to Challenge Nvidia's Reign

SEOUL, South Korea – March 30, 2026 – South Korean AI chip leader Rebellions has secured $400 million in a pre-IPO funding round, catapulting its valuation to approximately $2.34 billion and signaling a major escalation in the global race for AI infrastructure dominance. The funding, co-led by Mirae Asset Financial Group and the state-backed Korea National Growth Fund, coincides with the launch of the company's first fully integrated hardware systems, RebelRack™ and RebelPOD™, aimed directly at solving the efficiency and scalability crisis in production AI.

This capital infusion, which brings Rebellions' total raised to $850 million, with a staggering $650 million secured in the last six months alone, sets the stage for an aggressive push into the U.S. market and prepares the company for a future public offering. The move comes as the AI industry confronts a critical turning point: the challenge is no longer just creating powerful models, but deploying them economically and efficiently within the constraints of existing data centers.

A New Paradigm for AI Infrastructure

Rebellions is positioning itself as the answer to a market grappling with the immense power consumption and cost of traditional AI hardware. The company argues that the era of measuring AI by brute force alone is over, shifting the focus to practical, real-world inference—the process of running trained AI models to generate results.

"AI is now measured by its ability to operate in the real world — at scale, under power constraints, and with clear economic return," said Sunghyun Park, Co-Founder and CEO of Rebellions, in a statement. "That shifts the center of gravity toward inference infrastructure and software that makes that infrastructure usable."

At the heart of this strategy are the newly launched RebelRack™ and RebelPOD™. These are not just chips, but fully deployable, vertically integrated systems. The RebelRack™ is a production-ready unit of inference compute, while the RebelPOD™ integrates multiple racks into a scalable cluster for large-scale deployments. Both are built upon the company's flagship Rebel100™ Neural Processing Unit (NPU).

The Rebel100™ itself is a feat of engineering, employing a chiplet-based architecture with four compute dies manufactured by Samsung. It incorporates 144 GB of high-bandwidth HBM3e memory, delivering a petaFLOP of 16-bit floating-point performance. This design is purpose-built for inference, prioritizing performance-per-watt and cost-efficiency to allow organizations to operate powerful AI workloads without needing to build entirely new, power-hungry data centers.

South Korea's National Champion in the Global AI Race

The funding round is as much a strategic national statement as it is a corporate milestone. The participation of the Korea National Growth Fund, which chose Rebellions for its first-ever investment under the government's "K-Nvidia initiative," underscores the company's central role in South Korea's ambition to achieve AI and semiconductor sovereignty. This initiative is a direct effort to cultivate a domestic champion capable of competing with global giants and reducing reliance on foreign technology.

"Amid the global race for AI leadership, venture capital plays a critical role in identifying and backing high-potential companies," stated Eung-Suk Kim, Vice Chairman and CEO of Mirae Asset Venture Investment. "We are proud to support Rebellions – the first company backed by the Korean National Growth Fund – as a strategic partner in demonstrating its capabilities and value on the global stage."

This national backing, which follows the strategic merger of Rebellions and SK Telecom's AI chip unit Sapeon in 2024, creates a formidable entity. The support from investors like Samsung and SK Hynix also provides a crucial advantage by helping to secure a stable supply of advanced memory chips, a well-known bottleneck in the AI hardware industry.

An Open Ecosystem to Empower Developers

A key pillar of Rebellions' strategy is its deep commitment to open-source software, a direct counterpoint to the proprietary ecosystems that have historically dominated the hardware landscape. The company has built a cloud-native AI software stack based on open standards like Kubernetes, designed to work seamlessly with popular frameworks including vLLM, PyTorch, Triton, Hugging Face, and OpenShift.

This software-centric approach is designed to eliminate friction for developers and prevent vendor lock-in for enterprises. By aligning with tools that AI/ML teams already use, Rebellions aims to drastically shorten evaluation and adoption cycles. "The companies that succeed in this era will not be defined by silicon alone, but by how effectively they integrate into the open source software ecosystem and enable developers to build and deploy without friction," Park emphasized.

This strategy acknowledges that true scalability comes not just from powerful hardware, but from software that abstracts its complexity and allows for flexible, heterogeneous deployments across diverse environments. For enterprises, this means the ability to integrate more efficient inference hardware without overhauling their existing software stacks or retraining their teams.

The Push into America and Beyond

With a war chest of fresh capital and a mature product line, Rebellions is launching a focused expansion into the United States, led by newly appointed Chief Business Officer Marshall Choy. The company is targeting a market that is rapidly maturing beyond the simple need for more compute.

"The U.S. market is reaching a point where access to compute is no longer the only question — how efficiently that compute is used is becoming just as important," said Choy. "Organizations are looking for infrastructure that works within their existing environments, extends the life of their current investments, and enables new revenue-generating AI applications. That is where we are focused today."

Leveraging its proven track record with large-scale deployments for telecom operators like KT in South Korea, Rebellions is targeting U.S. government agencies, emerging "Neocloud" providers, and large enterprises. The company is already running proof-of-concept trials with American customers, aiming to demonstrate how its systems can deliver immediate returns on investment. With a clear path toward a future IPO, Rebellions is not just building chips, but a complete ecosystem to power the next wave of AI adoption.

"Our objective is to build a durable infrastructure company that enables the next phase of AI adoption," concluded Park. "That requires long-term investment in software, systems, and ecosystem — in addition to hardware — and a clear focus on making AI deployable at scale."