We were pleased to announce earlier this week that Deeplite was crowned as a Silver Stevie winner in the 2022 International Business Awards, in the Artificial Intelligence/Machine Learning Solution category. The team was thrilled to receive this honor and get recognition for Deeplite Neutrino and DeepliteRT from such a prestigious award program.
Neutrino and DeepliteRT’s computer vision technology were recognized for the way they seamlessly fit into an organization’s AI strategy to allow large AI models to work on edge devices, like security cameras and drones. This means small organizations and startups can harness the power of AI and take advantage of the edge. Deeplite is working with organizations in security and surveillance, manufacturing, automotive and more to develop computer vision solutions that help perform various tasks including identifying porch pirates and reading speed limit signs to make roads safer.
How it works
Neutrino is Deeplite’s core software, providing a unique suite of AI-focused functionality to support training-aware model optimization, two-bit quantization while reducing memory bandwidth and retaining model accuracy. It provides compute optimization that delivers the highest inference performance and power efficiency, and is the optimal deployment path for PyTorch vision models for Arm-based embedded systems.
The software is used on-premises to provide data privacy and security to customers. With flexible APIs, it can seamlessly fit into a company’s routine AI workflow with minimal effort and is designed to be integrated within existing AI frameworks. The end result is a simple and scalable way for computer vision and AI engineers to run their models seamlessly, cost effectively and to enable fast time-to-market for embedded device AI applications.
Going green with Deeplite
One of the many benefits of using Deeplite’s optimization software is the environmental impact, which was also recognized in the Stevie award. This is because the compute power - particularly in the cloud - required to develop, train and run machine vision AI is a growing concern, especially as companies become more aware of their carbon footprint. AI developers’ cloud-compute costs are escalating, and businesses are facing more questions from investors and customers about issues of privacy and carbon emissions. DeepliteRT requires less on-device compute power, thus creating a “greener” way to enable AI models on edge devices. In addition, Neutrino’s efficient, on-premises, automated software engine can transform inefficient but accurate AI models into efficient models for size, speed and power consumption in record time, thereby requiring less compute power.
On the awards front
This Stevie award adds to Deeplite’s recent list of accolades, which also include a Tech Trailblazer award for AI Trailblazer (check out Nick’s podcast interview on that here) and an AI Tech award for Deep Learning Technology. We were also named as a finalist for the Business Intelligence Group’s Artificial Intelligence awards, and we we’ve been named to CRN’s Hottest AI Startups list. In addition, insideBIGDATA’s has named Deeplite to its IMPACT 50 list for multiple quarters.
Interested in learning how Deeplite can optimize AI for your organization? Contact us today!