SEMIFIVE, a provider of custom AI semiconductor (ASIC) solutions, announced on September 29, 2026, that it has signed a contract worth approximately USD 52 million with a U.S.-based AI fabless semiconductor company to develop a next-generation AI inference accelerator.

The deal is the company's first "Spec Hand-off" engagement in the North American market, and the company describes it as its largest single contract to date.

A Contract of Significant Scale for the Company

According to SEMIFIVE, the contract by itself represents more than 40% of the company's total orders of USD 124.3 million in 2025. It also accounts for approximately 60% of the USD 87.7 million in new orders the company secured during the first half of 2026.

The company said the win strengthens its presence in North America while reinforcing its technological competitiveness in the global custom AI semiconductor market.

How the Spec Hand-off Model Works

Under SEMIFIVE's Spec Hand-off model, customers only need to provide their key performance requirements and specifications. SEMIFIVE then takes full responsibility for the entire development process, including detailed design, software development, packaging, testing, and mass production.

The company distinguished this approach from conventional turnkey engagements, in which customers typically complete the chip design before outsourcing subsequent implementation and manufacturing processes. In the Spec Hand-off model, SEMIFIVE takes the lead from an earlier stage of chip development.

SEMIFIVE noted that specialized ASIC companies are playing a broader role from the requirements stage across the global custom semiconductor market. As examples, it pointed to major technology companies including Google and Amazon, which have worked with semiconductor companies such as Broadcom and Marvell to develop their own AI silicon.

Technical Ambitions for the Inference Accelerator

For the project, SEMIFIVE will integrate advanced technologies intended 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 (TCO).

The accelerator will adopt LPDDR6, a next-generation low-power DRAM technology, to reduce memory-related power consumption. It will also use 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 also plans to implement an architecture designed to deliver both high compute throughput and operational stability.

Timeline and Target Market

Tape-out is targeted for the first half of 2027. After that milestone, the two companies plan to begin global mass production in 2028, targeting global hyperscalers and cloud service providers (CSPs).

Company Perspective on the Inference Market

Brandon Cho, CEO and Co-founder of SEMIFIVE, framed the deal in the context of growing demand for tailored silicon. "As the AI inference market continues to expand rapidly, demand is accelerating for custom silicon optimized for each customer's services and workloads," he said.

"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."

Recent Company Milestones

The announcement follows other recent news from SEMIFIVE. The company has said it began mass production of a data center LLM AI inference accelerator called "Bertha," developed for HyperAccel, on Samsung 4nm.

It also reported that it accelerated growth with a 97% year-over-year revenue surge in the first half of 2026, which it attributed to its ASIC turnkey business, mass production, and IP. In that report, the company cited bookings of USD 87.7 million, the same figure it references for first-half new orders in the latest announcement.

The identity of the U.S.-based fabless customer was not disclosed in the announcement.

From Fab Foundations to Full Production: Track Every Semiconductor Project Across South Korea

South Korea is racing to become the world's semiconductor powerhouse, with billions pouring into new fabs, packaging facilities, and R&D campuses in South Korea. But can you pinpoint which of these projects are moving forward, which are stalled, and where your next opportunity lies?

The Global Project Tracking (GPT) platform by Blackridge Research gives you real-time visibility into semiconductor capital projects across the region, so your business development, sales, and strategy teams never miss a signal.

Every project record is enriched with key details, project stage, owners and stakeholders, expected timelines, and equipment and services requirements, helping you engage the right decision-makers at the right moment.

• Upcoming Projects

• Tender Notices

• Contract Awards

• Projects Under Construction

• Completed Projects

Ready to see how leading semiconductor suppliers and contractors are building their South Korea pipelines? Book a Free Demo of the Global Project Tracking (GPT) platform by Blackridge Research today.