Cost effective deployment of vision-language models for pet behavior detection on AWS Inferentia2
Not a pilot. Tomofun deployed vision-language models across Furbo Pet Camera using AWS Inferentia2 to cut inference costs while maintaining accuracy.
Why it matters
Case study demonstrating real-world cost optimization for vision-language model deployment on purpose-built AI chips. Shows how startups are using AWS Inferentia2 to make computer vision applications economically viable at scale.
The key facts
10 to knowTomofun (Taiwan pet-tech startup) deployed vision-language models for pet behavior detection
Used AWS EC2 Inf2 instances powered by Inferentia2 chips
Focus: cost reduction + accuracy maintenance
Application: Furbo Pet Camera (remote pet interaction)
Use case: Pet behavior detection via vision-language models
Tomofun (Furbo Pet Camera) deployed vision-language models on AWS EC2 Inf2 instances
AWS Inferentia2 used for pet behavior detection inference
Cost reduction as primary driver for chip selection
Maintained accuracy while reducing deployment costs
Taiwan-based pet-tech startup scaling AI inference workloads
Go to the source
AWS Machine Learning Blogaws.amazon.com
Publisher excerpt: Tomofun, the Taiwan-headquartered pet-tech startup behind the Furbo Pet Camera, is redefining how pet owners interact with their pets remotely. To reduce costs and maintain accuracy, Tomofun turned to EC2 Inf2 instances powered by AWS Inferentia2, the Amazon purpose-built AI chips. In this post, we…