“After working closely with Deci’s team, I’m excited to see how their innovative technology will make the most out of any inference hardware in terms of performance and accuracy. They develop with a passion to enable DL practitioners to deliver insane results quickly and easily.”
Guy Boudoukh, Intel AI Research
“We are excited to be working with Deci's platform - it provided amazing results and achieved a 4.6X acceleration on a model we ran in production and helped us provide faster service to our customers.”
Amos Bercovich, Senior Algorithm Team Leader, WSC Sports
“Deci is the perfect choice for deep learning executives, enabling them to substantially reduce time-to-market by boosting their deep learning developers with superpowers. No more endless optimization cycles… their AutoML-based technology is the next generation of deep learning development.“
Guy Uziel, Co-founder and CTO, LitiGate
Make your deep learning models blazing fast (2X-10X) without compromising accuracy
Achieve more with less. Up to 80% reduction in cost-to-serve or energy usage
Cut development time from months to days with AutoML. No more endless experimentation and optimization cycles to reach production grade
Automatically optimize your models for cloud or edge, to fully exploit any hardware (CPU, GPU, FPGA, ASIC, etc.)
Why use manual optimizations for your deep learning models when you can automatically deliver production-grade models faster than ever? Our Automated Neural Architecture Construction (AutoNAC) engine autonomously redesigns your models to squeeze the maximum utilization out of the hardware and leverage the hidden structures of the data.
The outcome is an AutoNAC model at its performance sweet spot, ready for production use.
Automated quality assurance
Want to understand the technology behind the Deci platform?