Avalanche Computing, a hyper-scale computing technique provider based in Taipei and the Bay Area, has developed a one-click-through AI framework that allows for production speed and is highly scalable.
Avalanche Computing's AI framework effectively trains your AI model in parallel or distributed mode on multiple GPUs without changing the algorithm for reducing AI project costs, and then does AI model deployment from one to many edge devices in one command.
Avalanche Computing focuses on optimizing the deep learning environment and training in parallel, lightening the trained model, and then deploying the AI model at a hyper-scale. The proposed AI framework helps increase the speed of manufacturing and medical applications, including predictive maintenance, anomaly detection, machine diagnosis, defect detection, supply chain analysis, operation optimization, EGG signal, breath sound, medical signal, medical speech to text, anomaly detection, lung cancer detection, and more.
Turnkey Solution for AI Development and Deployment
"Our mission is to help our customers who focus on intelligence services and applications save time, staff numbers, and money during AI development. With our deep tech and computing domain knowledge, we aim to provide an excellent workflow for AI development and deployment in one platform at scale. Besides, our many alliance partners, academic collaborations, and computing hardware partners have further helped our AI customers succeed in global business," said Avalanche Founder and CEO Jay Chen.
"With our AI experience with global enterprises, we can help customers achieve high-speed AI training, reduce waiting time, and quickly scale the inference system without building a new engineer team."
Developing AI to Suit Different Needs across Different Industries
Chen said his company currently serves more than 15 enterprise-level manufacturing and medical AI companies and SMEs. Avalanche's focus markets include the USA, Japan, Taiwan, and a several European countries, with target sectors including smart manufacturing, Industry 4.0, smart medical, and more. He related that his company has worked globally with NVidia SA, cloud service providers, and many software partners as well as with two accelerators, including NTUTEC and NTPC-AWS Joint Innovation Center.
The global news network AVING News has dispatched reporters to CES for 15 consecutive years since 2006, reporting the most number of news globally. After exhibition, it will hold CES 2021 Review Conference to share insight on CES at the end of January.
* MIK (Made in Korea) project in CES 2019-2020 to promote Korean companies to the world * BEST OF CES since 2015 to select outstanding products by each sector * The first CES review event CES Review Conference held every year since 2015