SEMIFIVE Commences Mass Production of HyperAccel's LLM AI Inference Accelerator 'Bertha' on Samsung 4nm, Spurring Growth Momentum

• Secures initial mass production contract for HyperAccel's AI inference accelerator, Bertha, with follow-on purchase orders anticipated as services expand
• Delivers a complete turnkey solution for a large-area "Big Die" chip of more than 500 mm², validating cutting-edge process leadership
• Expands portfolio from security and HPC to data center AI, fueling mass production ramp-up following KRW 42.3 billion in H1 order intake

SEOUL, South Korea, Sept. 8, 2026 /PRNewswire/ -- SEMIFIVE, a leading global provider of custom AI semiconductor (ASIC) solutions, announced today that it has begun mass production of a data center AI inference accelerator for HyperAccel.

This milestone marks SEMIFIVE's first large-scale mass production project utilizing Samsung Foundry's 4nm advanced process node. Following the successful initial production contract with HyperAccel, mass production volumes are projected to expand steadily, driven by follow-on purchase orders (POs) aligned with HyperAccel's ongoing service rollout and expansion.

The chip is an accelerator optimized for AI inference workloads, including large language models (LLMs), and is a "Big Die" of more than 500 mm². As die size increases, managing power consumption, heat dissipation, and manufacturing yield becomes far more challenging. SEMIFIVE brought this demanding advaned-node project into successful mass production by delivering a complete turnkey solution spanning front-end design and verification through packaging, testing, and volume manufacturing supply.

This achievement marks a full-scale ramp-up in SEMIFIVE's growth. Following the mass production of the 'Wisenet 9' AI ASIC chip for Hanwha Vision's security cameras in Q3 of last year, a high-performance computing (HPC) AI chip for a Japanese customer in Q2 of this year, and the data center AI inference accelerator in Q3 of this year, SEMIFIVE has added another key mass-production program to its pipeline. This demonstrates SEMIFIVE's ability to build stable, recurring revenue beyond one-time NRE projects through mass production. By aligning with customers' long-term product roadmaps and diversifying its portfolio across various application domains and technical complexities, the company is strengthening both business continuity and earnings momentum.

Backed by this expanding production lineup, the company is experiencing rapid financial growth. SEMIFIVE secured KRW 42.3 billion in new mass-production orders in the first half of this year alone—nearly double its full-year order intake of KRW 21.2 billion for the entirety of last year. Quarterly order intake surged by 71% from KRW 15.6 billion in Q1 to KRW 26.7 billion in Q2, with overseas orders accounting for 45% of Q2 bookings, marking the full-scale activation of its global business pipeline. With a virtuous cycle now established, design wins are translating into mass-production revenue. The quality of order intake has also improved, supported by stable OEM volumes backed by secured end customers and high-value data center ASIC projects.

"As the AI landscape shifts from training to inference, demand for ASIC accelerators in data centers is surging," said CEO of SEMIFIVE. "This successful mass production is a significant milestone that validates our unmatched execution capabilities in advanced processes. By combining our customer's innovative architecture with SEMIFIVE's comprehensive turnkey capabilities, we seamlessly executed the entire journey from design to volume production for a large-area die of more than 500 mm². We are committed to carrying the strong mass-production momentum demonstrated in the first half through the remainder of the year."

View original content to download multimedia:https://www.prnewswire.com/news-releases/semifive-commences-mass-production-of-hyperaccels-llm-ai-inference-accelerator-bertha-on-samsung-4nm-spurring-growth-momentum-302871908.html

SOURCE SEMIFIVE

[ Back To TMCnet.com's Homepage ]