modal-serverless-gpu - Run ML workloads on Modal serverless GPUs
Provides guidance for running machine learning workloads, deploying web APIs, and scheduling batch jobs on Modal serverless GPUs.
Tags
Updated: 2026-09-20Capabilities
Typical Inputs
Typical Outputs
What this skill does
- Configure serverless GPU resources
- Deploy ML web endpoints
- Define custom container images
- Manage persistent storage volumes
- Manage secret credentials
- Schedule recurring jobs
- Execute parallel processing
Inputs
- Python script files
- Modal authentication credentials
- Secret key-value pairs
- Model artifacts and datasets
Outputs
- Deployed web endpoints
- Remote execution logs and results
- Persisted storage volume state
Requirements
- Python 3.x
- modal>=0.64.0 package
- Linux, macOS, or Windows platform
- Authenticated Modal account