pipelines

Kubeflow Pipelines SDK 2.18 adds local pipeline visualization, richer component authoring, and better control over task retries, caching, and namespaces. Python 3.9+ remains supported.

Highlights

• Preview pipelines locally. Export standalone HTML/Mermaid visualizations and use local watch mode to inspect pipeline structure without deploying a KFP server. #15064
• Preview component outputs. Supported Python components can automatically generate Dataset and Model previews for the 2.18 run-details UI. Privacy controls include opt-outs and schema-only previews. #15087
• Author richer components. Python components support Pydantic BaseModel inputs and outputs. The Kubernetes extension adds init containers and Dynamic Resource Allocation (DRA). #13995, #13651, #14243
• Control retries and caching. set_retry(policy=...) selects which failures trigger retries with Argo Workflows. set_caching_options(cache_key=...) allows custom task cache keys. #14655, #13762
• Use namespaces explicitly. Client methods support explicit namespaces for immediate and recurring runs and private-pipeline lookup. Run listing gains a server-side pipeline filter, and runs can be created from a pipeline ID without specifying a version ID. #14812, #14611, #14660, #14711

Bug fixes

• Correct pipeline-ID-only run creation and private-pipeline namespace lookup. #14711, #14611
• Preserve embedded artifacts reached through symlinks and expose the underlying cause of Kubernetes configuration errors. #13554, #13871
• Fix ParallelFor task naming and component-description parsing edge cases. #13707, #13619

Installation

Install the matching 2.18.0 packages on Python 3.9 or later:

python -m pip install kfp==2.18.0 kfp-pipeline-spec==2.18.0 kfp-server-api==2.18.0

For Kubernetes-specific pipeline features, also install:

python -m pip install kfp-kubernetes==2.18.0

The four distributions remain separate. Upgrading these packages does not upgrade your cluster or rebuild existing component images. Generic container components do not acquire a Python requirement merely because KFP executes them.

Compatibility and upgrade guidance

• Backend and Kubernetes support: use a matching 2.18 backend for the new runtime features. DRA additionally requires Kubernetes 1.31+ with DynamicResourceAllocation enabled and an installed DRA driver; the feature is GA and enabled by default in Kubernetes 1.34. Older KFP backends reject the new DRA platform field.
• Retry policies and custom cache keys: retry policy selection is Argo-specific. An omitted policy uses the deployment's Argo configuration; bundled 2.18 manifests use OnError. Custom cache keys replace automatic input, image, command, argument, and PVC-name contributions, so callers must account for changes that should invalidate cached results.
• Preview privacy: previews may contain sample data or model configuration. Review privacy settings before enabling them for sensitive outputs. Existing artifacts are not backfilled.
• Private package mirrors: generated component installatio

What's changed

This maintenance release updates Go, Python, frontend dependencies and runtime images while retaining SDK support for Python 3.9.

• Backport the Go trust-store fix from #15025, including Go 1.27.3 and updated builders, with additional security dependency updates in #15026.
• Fix the proxy and cache-deployer runtime images (#15036).
• Remove frontend server development dependencies from the runtime image (#15040).
• Pin CI pipeline SDK installation to the tested commit (#15041), quote release manifest sources (#15030), and preserve release image publication with read-only workflow defaults (#15027).

Security and compatibility

Qualification of the released commit rebuilt and scanned all 24 image/platform combinations with zero scanner-reported fixable image findings. The four SDK packages passed installation and compilation checks on Python 3.9–3.13.

Security: Release-image scans found no vulnerabilities with available fixes at qualification time. Python 3.9 remains supported with documented dependency exceptions; the tested Python 3.10–3.13 environments had no findings. See the security qualification details and remaining findings.

The latest image tags now point to 2.17.3. Use the explicit 2.17.3 tag to pin deployments to this release. See the SDK release for Python packages.

Full changelog: https://github.com/kubeflow/pipelines/compare/2.17.2...2.17.3

Release of:

• KFP SDK
• KFP Kubernetes
• KFP Server API
• KFP Pipeline Spec

To install the KFP SDK:

pip install kfp-pipeline-spec==2.17.3
pip install kfp-server-api==2.17.3
pip install kfp==2.17.3
pip install kfp-kubernetes==2.17.3

For changelog, see https://github.com/kubeflow/pipelines/blob/sdk-2.17.3/sdk/RELEASE.md.

What's Changed

Full Changelog: https://github.com/kubeflow/pipelines/compare/2.17.1...2.17.2