---
id: kfp-server-api
version: "2.17.0"
license: Apache 2.0
license_treatment: permissive
maintenance: active
---

# kfp-server-api — Kubeflow Pipelines API

License: permissive · Maintenance: active · Downloads: 3.2M/mo

## What it is and what it does

kfp-server-api is an auto-generated Python client for the Kubeflow Pipelines REST API. It provides typed Python bindings to interact with a Kubeflow Pipelines service deployed on Kubernetes, allowing you to programmatically create, list, run, and manage ML pipelines and experiments. The package depends on standard HTTP and date-handling libraries (urllib3, certifi, python-dateutil, six) to handle API communication.

This is a low-level client library—it exposes the raw API surface rather than a high-level SDK. It is typically used by tooling, integrations, or advanced users who need direct API access rather than the higher-level Kubeflow Pipelines SDK. The package is actively maintained and generated from the Kubeflow Pipelines API specification.

Use it for:

- Build custom tooling or integrations that need to interact with a Kubeflow Pipelines service programmatically.
- Automate pipeline management tasks (create, list, run, monitor) from Python scripts or applications.
- Integrate Kubeflow Pipelines into CI/CD systems or workflow orchestrators that require API-level control.
- Develop monitoring or auditing tools that query pipeline status and execution history from a Kubeflow cluster.

## Worth the install?

AI-flagged interpretation of the facts above — verify before relying.

Provides Python client bindings for the Kubeflow Pipelines API, enabling programmatic interaction with Kubeflow Pipelines services running on Kubernetes.

Yes, if you need programmatic access to a Kubeflow Pipelines service and are comfortable with a low-level API client. The high install friction and lack of declared Python version support suggest this is aimed at users already familiar with Kubeflow and Kubernetes. No known vulnerabilities and active maintenance are positive signals. Not recommended if you're looking for a high-level SDK—use the main Kubeflow Pipelines SDK instead.

## Install

pip install kfp-server-api
uv add kfp-server-api
poetry add kfp-server-api

## Installing kfp-server-api

Before you install: High install friction due to being a generated OpenAPI client with multiple HTTP/date-handling dependencies. Actively maintained with recent releases, though the package itself is a thin wrapper around the Kubeflow Pipelines service API.

License in practice: Apache 2.0 is permissive; you can use, modify, and distribute this package freely in commercial and private projects with minimal restrictions.

Quickstart:

pip install kfp-server-api

from kfp_server_api.rest import ApiClient
from kfp_server_api.models.api_pipeline import ApiPipeline

client = ApiClient()
# Configure client with Kubeflow Pipelines server endpoint

Requires a running Kubeflow Pipelines service (typically deployed on Kubernetes) to connect to; this package is a client library only.

Verify before relying:

- Whether this package is intended for end users or primarily for SDK/tooling developers building on top of Kubeflow Pipelines.
- Specific Python version requirements or compatibility constraints not declared in the package metadata.
- Whether high install friction reflects genuine dependency complexity or transient build issues.

## Package facts

- License: Apache 2.0 (permissive)
- Python support: unspecified
- Install friction: high
- Maintenance: active
- Downloads: 3.2M/month (top 5,000 on PyPI)
- Known vulnerabilities: none known

## Tags

kubeflow pipelines client, kubeflow api python, ml pipeline orchestration client, kubernetes ml workflow api, kubeflow server api bindings, kubeflow, ml-orchestration, kubernetes

[View on SkillFed](https://skillfed.io/packages/kfp-server-api) · [View on PyPI](https://pypi.org/project/kfp-server-api/)