With the emergence of deep learning on mobile and embedded devices, AI solutions must satisfy strict computing requirements, which often limits performance and prevents people from moving to these devices. Deeplite created Neutrino, an intelligent optimization engine for Deep Neural Networks (DNNs) deployed on cloud servers and edge devices. Neutrino helps AI experts automatically optimize high-performance DNN models to satisfy target computation constraints. The engine inputs large, initial DNN models and constraints to produce smaller, highly-efficient models that meet these constraints.
- Compression: 16 to 300x
- Speedup: 2 to 10x
- Power savings: 20 to 50%
Get Started Today!
With a CMC subscription, you are eligible to become a Deeplite community user with full use of Deeplite Neutrino for research purposes. You will get hands-on with automated neural network optimization, create highly compact deep learning models and hear about new releases before anyone else does!
By joining the Deeplite community and sharing your feedback, you will help AI researchers, AI engineers and non-AI experts alike to make their models faster, smaller and more energy-efficient!