BrainChip Tapes Its Akida AKD1500 Out on GlobalFoundries' 22nm Node, Proving Its Portability
Now proven on a new node, BrainChip's Akida platform is likely to become a tempting target for those adding AI smarts to new chip designs.
BrainChip, a specialist in artificial intelligence at the edge, has announced that its AKD1500 Akida chip has been successfully taped out on GlobalFoundries' 22nm fully-depleted silicon-on-insulator (FD SOI) process node — proving that the core technology is portable between fabricators.
"This is an important validation milestone in the continuing innovation of our event-based, neuromorphic Akida IP, even if it is fully digital and portable across foundries," explains Anil Mankar, co-founder and chief development officer at BrainChip. "The AKD1500 reference chip using GlobalFoundries' very low-leakage FD SOI platform, showcases the possibilities for intelligent sensors in edge AI."
Inspired by the operation of the human brain, the Akida AKD1500 processor operates on spiking neural networks — solving problems expressible as such in a fraction of the power draw than rival machine learning hardware. The company launched its first commercially-available chips in January last year, after releasing a limited number of development kits based on the Raspberry Pi or an Intel-based miniature PC the year before.
The news comes hot on the heels of the addition of support for Akida devices to the Edge Impulse platform, which company co-founder and chief executive officer Zach Shelby claimed at the time "will provide users with a powerful and easy-to-use solution for building and deploying machine learning models on the edge."
Proving that its design is portable will be key to wider adoption of BrainChip's Akida technology, allowing users to license the IP for inclusion in their own chip designs — as RISC-V pioneer SiFive announced in April last year — and allowing future generations of the chip to benefit from easy portability to more efficient process nodes as they become available.
More information on the Akida processor is available on the BrainChip website.