Ctrlk
ProductResourcesGitHubBook a demo
For the complete documentation index, see llms.txt. This page is also available as Markdown.

Overview

Reference for the ZenML and Kitaru Python SDKs and CLIs.

ZenML ships two Python SDKs, each with its own reference site:

  • ZenML — build, manage, and deploy production-ready ML pipelines.

  • Kitaru — ZenML's sibling project for recording, replaying, and improving AI agents: real runs get recorded as sessions, then replayed against your real code with one thing changed (sessions, replays, evaluators, cohorts, experiments), published at sdkdocs.kitaru.ai.

ZenML SDK

The ZenML SDK provides Python tools and interfaces for building, managing, and deploying production-ready machine learning pipelines. This documentation helps you get the most out of ZenML's features.

Cover

SDK Documentation

Complete API documentation for all ZenML modules, classes, and functions.

Cover

CLI Methods

Command-line tools for managing ZenML pipelines, stacks, and deployments.

Cover

ZenML Client

Python client for interacting with ZenML projects, repositories, and services.

Cover

Integrations

Connect ZenML with popular ML tools, orchestrators, and cloud services.

Kitaru SDK

Kitaru is ZenML's sibling project for recording, replaying, and improving AI agents: traces you can run, not just read. Its Python SDK and CLI reference are generated from the Kitaru source and published on a dedicated site.

SDK & CLI Reference

Complete API reference for the Kitaru Python SDK and command-line interface.

Kitaru Client

Programmatic interface to a Kitaru server: sessions, replays, evaluators, cohorts, and experiments.

ZenML Scarf
NextClient

Last updated

Was this helpful?