SEMIFIVE signed a $52 million AI chip development deal with a U.S. company, expanding custom ASIC solutions for specialized AI workloads.

SEMIFIVE, a leading provider of custom AI semiconductor solutions, has signed a contract worth approximately KRW 70.3 billion (~US$52 million) with a U.S.-based AI fabless company to develop a next-generation AI inference accelerator. The project marks SEMIFIVE's first "Spec Hand-off" engagement in the North American market and represents the company's largest single contract to date.

The contract alone exceeds 40% of SEMIFIVE's total orders from 2025 (KRW 168.4 billion) and accounts for approximately 60% of new orders secured during the first half of 2026 (KRW 118.9 billion), significantly strengthening the company's presence in North America while reinforcing its technological competitiveness in the global custom AI semiconductor market.

Under SEMIFIVE's Spec Hand-off model, customers provide only their key performance requirements and specifications, while SEMIFIVE assumes full responsibility for the entire development process, including detailed design, software development, packaging, testing, and mass production. This contrasts with conventional turnkey engagements, where customers typically complete chip design before outsourcing implementation and manufacturing. SEMIFIVE instead leads from an earlier stage of development.

Specialized ASIC companies increasingly play a broader role across the custom semiconductor market, from the requirements stage onward. Major technology companies including Google and Amazon have developed their own AI silicon through partnerships with semiconductor companies such as Broadcom and Marvell.

For this project, SEMIFIVE will integrate advanced technologies to deliver the high memory bandwidth and data processing performance required for large-scale AI model inference while reducing power consumption and total cost of ownership. The accelerator will adopt LPDDR6, a next-generation low-power DRAM technology, to reduce memory-related power consumption, along with PCIe Gen5 to minimize bottlenecks in high-volume data transfers. Drawing on its experience with multiple Big Die projects of up to 800 mm², SEMIFIVE plans to implement an architecture designed to deliver both high compute throughput and operational stability.

Following tape-out in the first half of 2027, the companies plan to begin global mass production in 2028, targeting global hyperscalers and cloud service providers.

"As the AI inference market continues to expand rapidly, demand is accelerating for custom silicon optimized for each customer's services and workloads," said Brandon Cho, CEO and Co-founder of SEMIFIVE. "By successfully commercializing this next-generation inference accelerator and improving the economics of large-scale AI models, we aim to establish SEMIFIVE as an essential custom silicon development partner for the global hyperscale infrastructure market."