Secures KRW 18 Billion Project for Next-Generation AI NPU, Solidifying Leadership in High-Performance ASIC Market.Underscores AI Semiconductor Design Competitiveness with Consecutive 4nm AI Chip Contracts.

Secures KRW 18 Billion Project for Next-Generation AI NPU, Solidifying Leadership in High-Performance ASIC Market Underscores AI Semiconductor Design Competitiveness with Consecutive 4nm AI Chip Contracts SEOUL, South Korea, March 13, 2026 /PRNewswire/ -- SEMIFIVE, a leading global provider of custom AI semiconductor (ASIC) solutions, announced today that it has signed a turnkey design contract worth KRW 18 billion (approximately USD 12.5 million) with an AI semiconductor fabless company. This contract marks a follow-on project with the customer and covers the development of a high-performance AI NPU based on Samsung Foundry's advanced 4nm (SF4X) process technology. The project will further strengthen SEMIFIVE's leadership in the rapidly growing AI semiconductor market and expand the company's portfolio of high-performance AI chip design projects. SEMIFIVE continues to strengthen its position in the AI semiconductor market with a strong track record in NPU design and mass production. This achievement follows a series of major projects, including ▲XCENA, a CXL-based 4nm (SF4X) project, and ▲HyperAccel, an LPU-based 4nm (SF4X) project optimized for Large Language Model (LLM) inference. These consecutive wins further validate SEMIFIVE's technical maturity and execution capabilities in the high-performance AI semiconductor sector. The newly secured project aims to commercialize a next-generation AI NPU optimized for on-premise environments, where AI computations are performed locally without reliance on external networks. Designed for real-time processing of massive data, such as high-resolution Vision AI and LLM inference, the chip will feature advanced memory interfaces to structurally eliminate data bottlenecks. By maximizing Power, Performance, Area (PPA) efficiency, SEMIFIVE expects to deliver an unparalleled AI accelerator that combines high performance with low power consumption. SEMIFIVE's competitive edge lies in its proprietary SoC design automation system and end-to-end engineering solutions that drastically improve ASIC development efficiency. The company takes full responsibility for the entire development lifecycle, from chip specification, logical and physical design to packaging, software development, prototyping, and mass production. This allows customers to focus entirely on their core AI architecture and service competitiveness. Notably, SEMIFIVE is currently managing large-die turnkey projects of over 800 mm² and has secured high-speed interface design capabilities essential for high-performance AI silicon. Combined with its experience in managing global semiconductor supply chains, SEMIFIVE provides greater reliability for the commercialization of complex AI ASICs. "This contract once again proves that SEMIFIVE's accumulated technical references and custom design capabilities represent the strongest competitive advantage in the market," said Brandon Cho, CEO and co-founder of SEMIFIVE. "Through collaboration with customers possessing innovative AI architectures, we will maximize synergies in high-performance AI chip development and firmly seize leadership in the custom semiconductor market as an AI ASIC specialist." Meanwhile, SEMIFIVE is accelerating its global expansion by broadening its customer base across the United States, China, Japan, India, and Europe, building on its domestic AI semiconductor design projects. As demand for high-performance AI ASICs continues to grow rapidly, the company aims to establish itself as a strategic partner in the global AI ecosystem and as a leading custom semiconductor design specialist.

Physicists at the University of Oxford show they can assemble Schrödinger’s cat-like superpositions from far more exotic quantum ingredients.

Selection marks the next phase of construction on Micron’s New York project, expected to generate 50,000 jobs.

OneAdvanced has launched what it bills as the UK’s first private, sovereign healthcare large language model trained on NHS primary-care data.