Isolated container environments for safely running AI-generated code, RL rollouts, and agent workloads at scale, with sub-second scheduling, support for 100k+ concurrent sandboxes, integrated storage, and full observability.
Modal is an AI cloud infrastructure platform that provides on-demand, elastic GPU compute across multiple clouds, purpose-built for AI workloads including LLM inference, model training, fine-tuning, and sandboxed code execution, with sub-second cold starts, autoscaling to 1,000+ GPUs, and per-second billing.