BrainChip, SiFive Announce Partnership for High-Performance, Low-Power On-Device Edge AI
Spiking neural network acceleration on one side and high-performance low-power RISC-V on the other could make for an edge AI match-up.
Spiking neural network specialist BrainChip has announced a partnership with RISC-V pioneer SiFive on artificial intelligence at the edge — planning to combine their efforts to provide chip designers combined IP.
"We are pleased to partner with SiFive and have the opportunity to have our Akida technology integrated with their market-leading product offerings, creating an efficient combination for edge compute," says Jerome Nadel, BrainChip's chief marketing officer. "As we expand our ecosystem of portfolio partners, we want to be sure that these relationships are built on complementary technologies, enabling capabilities, and breadth of environments so that we can expand opportunities to as many potential customers as possible."
"Employing Akida, BrainChip’s specialized, differentiated AI engine, with high-performance RISC-V processors such as the SiFive Intelligence Series is a natural choice for companies looking to seamlessly integrate an optimized processor to dedicated ML accelerators that are a must for the demanding requirements of edge AI computing," adds Chris Jones, vice president of products at SiFive. "BrainChip is a valuable addition to our ecosystem portfolio."
BrainChip's Akida technology is claimed to dramatically accelerate a range of problem domains at a very low power draw through the use of spiking neural networks — operating in much the same way as the human brain. Earlier this year it announced it would begin selling Akida accelerators to the mass market for the first time, after having launched Raspberry Pi- and Intel-based development kits three months earlier.
SiFive, meanwhile, is a pioneer in the commercialization of the free and open-source RISC-V instruction set architecture, having launched both its own application-specific integrated circuits (ASICs) as well as making its IP — and that of its ecosystem partners — available for chip designers to integrate in custom parts. Two years ago the company turned its focus on edge AI with the Intelligence generator, which added the ability to define cores with the optional RISC-V Vector Extensions (RVV) for improved performance.
Thus far, however, the two companies have not confirmed any plans to release silicon combining their respective IPs — but both will be made available in SiFive's chip design ecosystem.
Freelance journalist, technical author, hacker, tinkerer, erstwhile sysadmin. For hire: freelance@halfacree.co.uk.