GLOBALFOUNDRIES Says it has qualified its 12LP+ finfet process and is ready for production.

GF’s differentiated 12LP+ solution is optimized for AI training and inference applications.

12LP+ introduces new features including an updated standard cell library, an interposer for 2.5D packaging, and a low-power 0.5V Vmin SRAM bitcell that supports the low latency and power-efficient shuttling of data between the AI processors and memory. The result is a semiconductor solution engineered to meet the specific needs of the fast-growing AI market.

12LP is being used by  Enflame, Tenstorrent, and other companies for AI accelerator applications.

12LP+ is claimed to have a 20% SoC-level logic performance boost over 12LP, and a 10% improvement in logic area scaling.

These advances are through its new standard cell library with performance-driven area optimised components, single Fin cells, a new low-voltage SRAM bitcell, and improved analog layout design rules.

12LP+, a specialized application solution, is augmented by an AI design reference package, packaging, and post-fab turnkey services.

The  IP validations for 12LP+ will be expanded  to include PCIe 3/4/5 and USB 2/3 to host processors, HBM2/2e, DDR/LPDDR4/4x and GDDR6 to external memory, and chip-to-chip interconnect for designers and clients pursuing chiplet architectures.

12LP+  has been qualified and is currently ready for production at  Fab 8 in Malta, New York.

Several 12LP+ tape-outs are scheduled for the second half of 2020.

Fab 8 will be made compliant with both the U.S. International Traffic in Arms Regulations (ITAR) standards and the Export Administration Regulations (EAR).

These new control assurances, which will go into effect later this year, will make confidentiality and integrity protections available for defense-related applications, devices, or components manufactured at Fab 8.