Modal

What is Modal?

Modal is a serverless cloud platform built specifically for AI and ML workloads. Define your compute environment in Python, and Modal handles infrastructure — provisioning GPUs, scaling up for batch jobs, scaling to zero when idle, and managing dependencies. No Dockerfiles, no Kubernetes, no infrastructure management. Run fine-tuning jobs on A100s, serve models with auto-scaling endpoints, schedule batch inference, and build data pipelines — all from pure Python code. Modal's cold start times are measured in seconds, not minutes. Pricing is per-second of compute with no idle costs. Used by AI startups and research teams who want powerful infrastructure without DevOps overhead.

Key Features

Frequently Asked Questions