Quadric, a company that jumped into the AI edge processing fray last year with a high-performance chip that it said didn’t require high-power CPU clusters, has raised $21 million to help fund its work on the next version of that chip.
The Burlingame, California, company’s ideas apparently are gaining popularity among Japan’s semiconductor players. The Series B funding round was led by NSITEXE, a Japan-based semiconductor firm and group company of mobility supplier DENSO, along with a “major” investment from Japanese fabless semiconductor company MegaChips. Existing investors Leawood VC, Pear VC, Uncork Capital, and Cota Capital also participated in the round.
The investment will enable the company to release the next version of their novel processor architecture, according to a Quadric spokeswoman, who added that the second-generation architecture will improve top-line performance, consume less power, and maintain the familiar M.2 form factor typical of expansion cards for developer use and deployments. The incoming money also will help Quadric improve its SDK performance, and roll out intellectual property products for system-on-a-chip integration.
While work proceeds on the company’s second-generation architecture, Quadric declined to name companies already using the first-generation offering. Samples of second-generation products will be available at the end of this year, the company said.
The funding news comes less than a year after Quadric announced its initial processor architecture, which touted the ability for edge processing of both older image process algorithms and newer AI and machine learning algorithms. At the time, Jack Gold, principal analyst at J. Gold Associates, told Fierce Electronics that so many companies were focusing on edge AI processing that some market shakeout is inevitable.
Quadric also has argued that while most other solutions combine high-power CPU clusters with application-specific NPUs, its own unified architecture is flexible enough to accelerate the entire application pipeline without the need for a powerful CPU.
“It’s an exciting time to be able to partner with industry powerhouses in semiconductors and AI to bring accelerated computing to the Edge,” said Veerbhan Kheterpal, Co-founder and CEO of Quadric. “The market is saturated with rigid accelerators. Our product fills the void with a fully programmable multi-kernel processing architecture.”
“As the AI market matures, businesses are under pressure to accelerate machine learning initiatives and need production-ready, easy-to-deploy AI solutions that don’t require building from the ground up," said Tony Cannestra, Director of Corporate Ventures at DENSO. “Having evaluated Quadric’s q16 processor, its ability to run many types of algorithms efficiently and flexibly allows Quadric’s platform to enable AI in new services and products.”
Quadric also plans to use some of the funding to help it expand its staff.
RELATED: Meta, with metaverse in mind, unveils AI Research SuperCluster