Tomofun Deploys Vision-Language Models on AWS Inferentia2 for Pet Behavior Detection

Pet-tech startup Tomofun uses AWS Inferentia2 chips to reduce costs while maintaining accuracy in its Furbo Pet Camera's AI detection capabilities.

According to Amazon AWS AI, Tomofun, a Taiwan-based pet-tech startup behind the Furbo Pet Camera, has adopted AWS Inferentia2 chips to optimize the deployment of vision-language models for pet behavior detection. The company turned to EC2 Inf2 instances powered by AWS Inferentia2, Amazon’s purpose-built AI chips, to achieve cost-effective inference while maintaining detection accuracy.

The Furbo Pet Camera enables pet owners to monitor and interact with their pets remotely. By leveraging AWS Inferentia2, Tomofun aims to reduce operational costs associated with running AI models for pet behavior analysis. The move represents a practical application of specialized AI hardware for edge computing use cases in consumer products.

AWS Inferentia2 chips are designed specifically for AI inference workloads, offering an alternative to general-purpose computing resources. According to the source, this deployment strategy allows Tomofun to balance cost efficiency with the performance requirements of vision-language models used in real-time pet monitoring applications.