Kubeflow Components
Kubeflow Pipelines (KFP) is a platform for building then deploying portable and scalable machine learning workflows using Kubernetes.
Kubeflow Notebooks lets you run web-based development environments on your Kubernetes cluster by running them inside Pods.
Kubeflow Central Dashboard is our hub which connects the authenticated web interfaces of Kubeflow and other ecosystem components.
Kubeflow Trainer is a Kubernetes-native project for fine-tuning LLMs and model training. It enables scalable and distributed training with various frameworks including PyTorch, JAX, TensorFlow, and more.
Katib is a Kubernetes-native project for automated machine learning (AutoML) with support for hyperparameter tuning, early stopping and neural architecture search.
KServe (previously KFServing) solves production model serving on Kubernetes. It delivers high-abstraction and performant interfaces for frameworks like Tensorflow, XGBoost, ScikitLearn, PyTorch, and ONNX.
Kubeflow Model Registry is a cloud-native component that provides a single pane of glass for ML model developers to index and manage models, versions, and ML artifacts metadata. It fills a gap between model experimentation and production activities. It provides a central interface for all stakeholders in the MLOps lifecycle to collaborate on ML models.
