Fujitsu to Launch 2nm 3D-Stacked MONAKA CPU and Made-in-Japan Server for Sovereign AI Infrastructure

Fujitsu Limited has announced that it will begin global sales of its next-generation CPU, FUJITSU-MONAKA, starting November 2026, alongside the launch of the made-in-Japan Fujitsu MONAKA Server. The move positions the Japanese technology giant at the forefront of a growing global push toward sovereign AI infrastructure, as governments and enterprises increasingly seek to retain control over the computing platforms that underpin artificial intelligence workloads.

The FUJITSU-MONAKA CPU is designed and developed entirely in Japan and will be sold both as a standalone processor to cloud and data center operators and server vendors, and as the engine powering the Fujitsu MONAKA Server. The server will be made broadly available from November 2026 to data center operators, enterprises, and the academic and HPC sectors in Japan and Europe, as well as to the defense sector, where supply chain integrity and operational autonomy are paramount.

At the heart of the announcement is a processor that integrates Fujitsu's extensive expertise in processor and server development. The FUJITSU-MONAKA utilizes a 3D-stacked architecture combining a state-of-the-art 2nm process for the core with a 5nm process for the cache and I/O sections, achieving a maximum operating frequency of 3.8GHz and memory transfer speed of 8800MT/s. By pairing dedicated hardware acceleration instructions for matrix operations with SVE2 vector operations and software optimization, the CPU delivers twice the AI inference throughput of comparable processors while effectively halving the number of servers and power consumption required for equivalent processing loads.

The strategic significance of domestic manufacturing runs throughout the announcement. The Fujitsu MONAKA Server is designed, developed, and manufactured in Japan, with production taking place at Fujitsu's Kasashima Plant. This ensures traceability by making the origin of components and manufacturing history visible and trackable, strengthening supply chain transparency and reliability at a time of rising geopolitical risk. The server also features confidential computing capabilities that protect data and applications running in memory with hardware encryption, supporting advanced AI utilization in secure environments, including multi-tenant cloud deployments.

Fujitsu is positioning the server as a solution to two of the most pressing challenges in AI infrastructure expansion: power consumption and physical space. The 1U server model adopts CDI/CXL technology, allowing for optimal allocation of memory and accelerators beyond the server's physical boundaries, mitigating memory shortages that commonly constrain AI inference. Leveraging Fujitsu's long-cultivated cooling technologies, the server can operate in ambient temperatures up to 40 degrees Celsius with air cooling and water temperatures up to 45 degrees Celsius with water cooling, without requiring specialized cooling infrastructure, reducing server cooling power consumption by up to 80 percent.

Beyond the hardware, Fujitsu will deploy a vertically integrated sovereign AI model with the MONAKA Server at its core, combining its proprietary AI technologies—including the AI platform Fujitsu Kozuchi and enterprise-grade generative AI Takane—with industry-specific models. This provides a sovereign AI platform with enhanced measures for traceability, security, and operational autonomy transparency.

Fujitsu plans to provide FUJITSU-MONAKA to cloud and data center operators and server vendors starting in the fourth quarter of fiscal year 2026, with global availability including the US and APAC regions. The Fujitsu MONAKA Server will be offered to select customers in the financial, telecommunications, and manufacturing sectors starting in the second half of fiscal year 2026, with shipments commencing sequentially from April 2027 in Japan and Europe. The company also plans to add high-density servers for AI data centers and rack-scale servers enabling autonomous operation to its product lineup.