Deci raises $9.1M to optimize AI fashions with AI

Deci, a Tel Aviv-based startup that’s constructing a brand new platform that makes use of AI to optimized AI fashions and get them prepared for manufacturing, at this time introduced that it has raised a $9.1 million seed spherical led by Emerge and Sq. Peg.

The final thought right here is to make it simpler and sooner for companies to take AI workloads into manufacturing — and to optimize these manufacturing fashions for improved accuracy and efficiency. To allow this, the corporate constructed an end-to-end answer that permits engineers to usher in their pre-trained fashions after which have Deci handle, benchmark and optimize them earlier than they package deal them up for deployment. Utilizing its runtime container or Edge SDK, Deci customers may then serve these fashions on nearly any fashionable platform and cloud.

Deci’s insights display combines all indicators of a deep studying mannequin’s anticipated habits in manufacturing, ensuing within the Deci Rating – a single metric summarizing the general efficiency of the mannequin.

The corporate was co-founded by co-founded by deep studying scientist Yonatan Geifman, know-how entrepreneur Jonathan Elial, and professor Ran El-Yaniv, a pc scientist and machine studying skilled on the Technion – Israel Institute of Technology.

“Deci is main a paradigm shift in AI to empower knowledge scientists and deep studying engineers with the instruments wanted to create and deploy efficient and highly effective options,” says Yonatan Geifman, CEO and co-founder of Deci. “The quickly rising complexity and variety of neural community fashions make it onerous for firms to attain high efficiency. We realized that the optimum technique is to harness the AI itself to sort out this problem. Utilizing AI, Deci’s purpose is to assist each AI practitioner to resolve the world’s most advanced issues.”

Deci’s lab display allows customers to handle their deep studying fashions’ lifecycles, optimize inference efficiency, and put together fashions for deployment. Picture Credit: Deci

The corporate guarantees is that, on the identical {hardware} and with comparable accuracy, Deci-optimized fashions will run between 5 and ten occasions sooner than earlier than. It might make use of CPUs and GPUs for working its inference workloads and the corporate says that it’s already working with prospects in autonomous driving, manufacturing, communication and healthcare, amongst others.

“Deci‘s potential to routinely craft top-performing deep studying options is a paradigm shift in synthetic intelligence and unlocks new alternatives for a lot of companies throughout completely different industries,” stated Liad Rubin, Companion at Emerge. “We’re proud to have partnered with such unbelievable founders and be a part of Deci’s journey from day one.”


What do you think?

Written by Sourov


Leave a Reply

Your email address will not be published. Required fields are marked *



Main Edge Tools has a know-how to enhance photo voltaic manufacturing and $7.6 million to go to market

Uber’s ‘robo-firing’ of drivers focused in newest European lawsuit