⚡ Modal

Modal is a serverless cloud platform designed for AI and ML workloads. It lets you run GPU inference, batch processing, and scheduled jobs in Python without managing any infrastructure — defining compute requirements as decorators on plain Python functions.

• GPU inference with @app.function(gpu=...) — A10G, A100, H100 options
• Modal.cls Model class for cached model weights across calls
• Volumes for persistent model weight storage across cold starts
• Parallel batch processing via .map() and .starmap()
• Web endpoints, cron scheduling, and Secrets management