Partnerships
Preferred Networks Targets Faster AI Inference with MN-Core L2000

Key PointsMemory bandwidth is the real bottleneck for inference because generating output…
View more
- Memory bandwidth is the real bottleneck for inference because generating output requires reading hundreds of gigabytes of data.
- SEMIFIVE handles design implementation and supply-chain coordination for the L2000's 3D-IC approach.
- Power consumption heat dissipation and manufacturing yield make 3D-stacked AI chips harder to develop than conventional chips.
Preferred Networks and SEMIFIVE announced on November 16, 2026 that they are collaborating on the MN-Core L2000 Series, AI inference chips designed to deliver more than twice the memory bandwidth of PFN’s L1000 Series.
The L2000 succeeds the L1000, which is scheduled for release in 2027. Work on the newer series is already underway, with tape-out planned for the first quarter of the same year. Tape-out is the completion of the chip design and handoff of final design data to the manufacturer.
Why Memory Bandwidth Is Central
PFN’s November 2024 L1000 announcement explains the design priority. Training generative AI demands substantial computing resources, but inference requires the processor to read hundreds of gigabytes of data whenever it generates a unit of text, images or other output. Limited memory bandwidth can therefore become a bottleneck even though inference needs fewer computing resources than training.
Both L-series designs stack DRAM vertically with logic chips based on MN-Core. PFN says this 3D structure achieves substantially higher memory bandwidth than GPUs and other AI processors, supporting faster inference and better power efficiency for applications including AI agents. The L1000 design likewise combines densely mounted arithmetic units with distributed memory, both controlled by software to limit power consumption and heat.
PFN has been developing MN-Core since 2016, and says supercomputers using the first-generation architecture topped the Green500 energy-efficiency ranking three times. That history applies to the earlier architecture, while the L2000’s bandwidth improvement remains a development target.
Three Companies Share Development Work
PFN is responsible for the L2000’s processor architecture and logical design. SEMIFIVE, a turnkey provider for custom semiconductors, is handling design implementation and supply-chain management through its 3D-IC design solution. Its coordination spans DRAM supply, volume manufacturing, advanced packaging and testing.
SEMIFIVE CEO Brandon Cho described the project as the company’s “first full-scale 3D-IC initiative.” The assignment places SEMIFIVE between PFN’s processor design and the specialized memory, packaging and manufacturing operations needed to implement it.
Samsung Foundry’s platform technology will be used to manufacture the logic chip. Samsung Foundry is also providing 3D-IC design infrastructure, including optimized I/O interfaces that connect the stacked dies and design methods for optimizing the power structure.
PFN co-founder and CEO Daisuke Okanohara identified power consumption, heat dissipation and manufacturing yield as challenges that make 3D-stacked AI chips more complex to develop than conventional chips. Those issues are significant because the performance approach depends on integrating logic and memory vertically rather than fabricating a standalone logic device.
Government Support and Next Steps
The L2000 development is supported by Japan’s Acquisition, Technology & Logistics Agency under its Innovative Science and Technology Initiative for Security program. The support comes through a Type S project running from fiscal 2024 through fiscal 2028.
PFN and SEMIFIVE also plan to jointly showcase the L2000 at PFN’s booth during SC26 in Chicago from November 17 to 19, 2026.
The partnership gives PFN an implementation and manufacturing path for its bandwidth-focused architecture and broadens SEMIFIVE’s 3D-IC work. The central execution test is whether the companies can carry the design’s bandwidth and efficiency goals into production while controlling the power, heat and yield challenges Okanohara identified.












