Create a single system of record for ML models that brings ML/AI development in line with your existing SSC. JFrog Artifactory now natively supports ML Models including the ability to proxy Hugging Face, a leading model hub. Manage your ML models and all their associated files alongside the PyPi packages, Conan Libraries, and Docker Images that will ultimately comprise your AI enabled application.
Hugging Face and JFrog
Remote Hugging Face Repositories
Local Hugging Face Repositories
JFROG ARTIFACTORY
ML Model Scanning
Governance
Remote Hugging Face Repositories
Cache Open Source models from Hugging Face in a JFrog Artifactory Remote Hugging Face repo, protecting the models you rely on from deletion or modification.
Local Hugging Face Repositories
Store and manage proprietary and internally-augmented ML and GenAI models with scalability to support massive binary files
ML Model Scanning
Detect and block malicious models and those with non-compliant licenses.
Governance
Full control and governance over which models are brought into and leveraged by the ML/data science organization in anticipation of inevitable regulations to come
What file formats does JFrog support for models?
JFrog supports all file formats (i.e. pkl, onnyx) and frameworks as a derivative of Hugging Face.
Do JFrog Hugging Face local repositories support federation?
Yes, the model and data files will be included in federations.
Does JFrog act as a machine learning model repository or registry?
JFrog serves as a ML Model registry with all the benefits of Artifactory and Xray applied to an organization’s models.
