Modal

Modal is an AI infrastructure platform designed for developers to run inference, training, and batch processing with sub-second cold starts, instant autoscaling, and a local-like developer experience. It offers programmable infrastructure defined in code, built for performance with rapid container scaling, elastic GPU capacity across clouds, and unified observability. Modal powers various ML workloads including inference for LLMs and generative models, model training, sandboxes for untrusted code, and large-scale batch processing, all built on an AI-native runtime that is significantly faster than Docker.