Hyperparameter Tuning
User guides to run Hyperparameter Tuning Experiments
User Guides
User guides for Katib
Deprecated
The Katib project is deprecated. Users should migrate to the OptimizationJob APIs, which are released as part of the Kubeflow Trainer project under alpha feature gate.
OptimizationJob is the next step for hyperparameter optimization in Kubeflow. It is natively integrated with the TrainJob API, so every trial can use Trainer runtimes and distributed training features out of the box. Metrics are pushed through the Kubeflow SDK and Trainer status server, so you no longer need a separate database or a metrics collector sidecar.
Work on the remaining features is tracked in kubeflow/trainer#3562.
How to Optimize Hyperparameters for LLMs Fine-Tuning with Kubeflow
API description of LLMs Hyperparameter Optimization using Katib
Neural Architecture Search
User guides to run Neural Architecture Search Experiments
How to use Katib UI
How to access and use Katib UI
How to use Trial Templates
Trial template parameters overview and how use CRDs with Katib Trials
How to Configure Metrics Collector
Overview of Katib metrics collector and how to configure it
How to Configure Early Stopping
Early Stopping overview for Katib Experiments
How to Resume Experiments
How to modify running Experiments and resume completed Experiments
How to configure env variables
List of environment variables accepted by Katib components
How to use Katib Config
Katib configuration overview and how to update values
Katib Installation Options
Overview of the ways to install Katib control plane
Feedback
Was this page helpful?
Thank you for your feedback!
We're sorry this page wasn't helpful. If you have a moment, please share your feedback so we can improve.
Last modified November 6, 2026: katib: Announce Katib Deprecation (#4519) (af17adf0)