The unified framework for building, shipping, and scaling production AI applications and fast model inference APIs.
Core packaging, containerization, and inference serving framework is 100% open source; BentoCloud includes onboarding credits.
BentoML is the open-source framework for building reliable, scalable, and cost-efficient AI applications. It standardizes model packaging, dependency management, and adaptive batching into reproducible Bento artifacts, ready to deploy across Kubernetes, Docker, or serverless BentoCloud.