To accelerate our industry’s path to an open standard, we have submitted KitOps to the CNCF so that others can more easily contribute to it, and benefit from it.
KitOps is a packaging, versioning, and sharing system for AI/ML projects, using open standards to work seamlessly with your existing AI/ML, DevOps, and development tools, all stored in your enterprise container registry.
KitOps generates a ModelKit for your AI/ML project, including everything needed for local reproduction or production deployment. ModelKits are immutable, signable, and live in your registry, making them easy to track, control, and audit.
ModelKits simplify collaboration between data scientists, developers, and SREs by allowing selective unpacking to save time and space. Teams use KitOps for secure, efficient AI/ML project management across the lifecycle.
Use KitOps for all AI/ML projects:
Predictive models
Large language models
Computer vision models
Multi-modal and audio models, etc.
As an OCI-compliant packaging format, ModelKit encapsulates datasets, code, configurations, and models into a single, standardized unit. This approach not only streamlines the development process but also ensures broad compatibility and integration with a vast array of tools and platforms.
AI is everywhere, but security and privacy remain major concerns. Tools that leverage LLMs offline and maintain strong context awareness can offer the best solution for assisting users while safeguarding their privacy.