Nota Optimizes AWS AI Chips: Lightweighting for Efficiency

Nota is launching a new service to optimize customer AI models for Amazon Web Services' in-house AI chips. This tuning service is designed to lightweight AI models, making them more efficient. Nota, an AWS certified partner, will use its Netspresso platform for this. Netspresso can reduce an AI model's size by over 90% while keeping its accuracy. The service includes technology verification, model porting, lightweighting, and performance tuning. It targets corporate customers who may not have AI optimization experts. Nota says this service will help improve cost efficiency and inference performance for AI models running on AWS AI chip environments like Trainium and Inferentia. This means businesses can get better performance from their AI at a lower cost.