Modal

Modal is a serverless cloud platform designed for running compute-intensive workloads, particularly AI and ML tasks. It provides on-demand access to GPUs and simplifies deploying Python functions that scale automatically.

Key features include: instant GPU access (A100, H100, etc.), pay-per-second billing, simple Python decorators for deployment, fast cold starts, and persistent storage. Modal handles container building, scaling, and orchestration automatically, letting engineers focus on code rather than infrastructure.

For AI engineers, Modal is valuable for: running inference on open models, fine-tuning jobs, batch processing, and any workload needing GPUs without managing infrastructure. The developer experience is remarkably simple. compared to traditional cloud GPU provisioning. Use cases include: deploying custom models, running periodic jobs, and building AI applications that need burst compute capacity.