We lift the lid of Deep Learning models
We make any deep learning model meet your requirements of:
- Hardware Cost
- Energy Consumption
- Performance (predictions/s)
We achieve this by a combining our proprietary algorithms with the latest innovations from the research community, so that you can rest assure that your application is always using state-of-the-art technology. More specifically, our toolchain includes methods for pruning, quantization, compression, effective memory usage and energy-aware computations.
% cheaper hardware
Times less Energy
times less dev time
Deep learning Optimisation
We have combined our propitiatory algorithms for efficient AI with state of the art algorithms from the research community. The combination is 4-16x faster and 5-15x more energy efficient inference.
There is a sea of offerings today for deep learning acceleration. How good are they, and more importantly, what offering is best for your specific deep learning model and application requirements? We will help you find the optimal hardware choice between CPU, GPU, FPGA or ASIC.
Every application is unique. We can help you with a customised rugged PCB design for the selected computational target and the needed interfaces to the rest of your system and sensors.
Pilots and Supporters
Are You currently considering embedding AI into your products?
MobilityXlab is a collaboration between Ericsson, Volvo Trucks, Volvo Cars, CEVT and Veoneer to promote collaborations with interesting startups. We have been part of MobilityXlab since its inception via our parent company Machine Intelligence Sweden AB.
Subscribe For Updates
We are happy to discuss your needs. Just reach out to see how we could be of service to you!
This project has received funding from the European Union’s Horizon 2020 research and innovation programme under grant agreement No 780681.
Mässans gata 10, 412 51 Gothenburg, Sweden
Box 5243, 402 24 Gothenburg, Sweden
EmbedlOpt Research and Technology AB is a spin-off from a collaborative research project on low energy computing, funded by the European Commission, where Machine Intelligence Sweden AB choose to spin-off of their research results into a new venture focused on helping companies in their journey of bringing prototype deep learning models to production.