Modal

Serverless GPU cloud for ML

0 case studies

1 specialists

General Infrastructure

Hire a Modal expert Visit modal.com

What it's used for

Running Python functions on cloud GPUs without managing any infrastructure — you define your environment in code and Modal handles container builds, scaling, and GPU scheduling. It's popular for batch inference, fine-tuning jobs, and deploying model endpoints where you want to pay per-second of compute rather than reserving instances.

Getting started

Install the modal Python package with pip install modal, then run modal setup to authenticate via your browser. Write a Python script using Modal's @app.function decorator with gpu="A100" (or similar) and deploy with modal deploy. Free tier includes limited GPU hours to get started.

 $ pip install modal 

Commonly paired with

Hugging Face

Weights & Biases

No case studies yet

Be the first to share a Modal case study and get discovered by clients.

Submit a case study

For hire

Modal specialists

Hire now

Charles Frye

Developer Educator @ Modal (ex-Full Stack Deep Learning)

Developer Educator at Modal, previously co-instructor of the Full Stack Deep Learning course at UC Berkeley. Creates in-depth tutorials on deploying LLMs and running GPU workloads. PhD in neuroscience.

ML EducationLLM DeploymentGPU Infrastructure +1

Available 53 projects San Francisco

$200/hr

Thought leaders

AI leaders using Modal

Follow for insights, tutorials, and thought leadership

Erik Bernhardsson

Modal Labs

Founder of Modal Labs, the high-performance serverless cloud for developers that reached unicorn status ($1.1B valuation) in September 2025. Previously spent 7 years at Spotify where he built the music recommendation system and created Luigi, the popular workflow scheduler. Also built Annoy, one of the first open-source vector databases based on ANN search.