Modelpack: Standardizing The Packaging And Distribution Of...
summary
ModelPack standardizes packaging by packaging AI/ML models as OCI artifacts and publishing them to any container registry; it ships a modelctl CLI that assembles model files, dependencies, and metadata, signs artifacts and generates SBOMs, and integrates with Jenkins/GitHub Actions for CI/CD. It reuses OCI tooling so registries, ORAS, and Harbor provide versioning, access control, and distribution; runtimes consume artifacts via Docker/Podman, Kubernetes image volumes or a CSI driver, and the project is a CNCF Sandbox since May 2025.
AI-written summary · about
related talks
05mProject Lightning Talk: ModelPack: Bringing the OCI Standard to AI/ML
41mModelPack: Packaging ML models as OCI artifacts made easy
23mBringing Container-Native Simplicity to AI/ML
18mBaby Sitting Your AI Models in Production? STOP!
34mOpen Standards For Machine Learning Model Deployment
20mBuilding Governable ML Models with R
41mOpen standards for machine learning deployment
46mProduction Model Deployment