We’re excited to announce that Python AI Toolkit for IBM z/OS version 1.14 is now available! Start exploring today and unlock new possibilities for AI on z/OS.

This latest release brings important enhancements designed to help developers and enterprises build, scale, and modernize AI and machine learning workloads on z/OS. With support for the latest Python runtime 3.14 and a significantly updated package ecosystem across python versions 3.12, 3.13 and 3.14, it is easier than ever to bring modern AI capabilities to mission-critical environments.

What’s New in Release 1.14 ?

This release focuses on platform modernization, dependency currency, and expanded ecosystem support, helping you build and run AI workloads more efficiently on z/OS.

• Modern Python Support (3.14)

  Python AI Toolkit now supports Python 3.14, enabling you to:
  • Use the latest Python language features.
  • Benefit from improved performance and runtime efficiency.
  • Stay aligned with the evolving Python ecosystem.

• Transition from Python 3.11

Support for Python 3.11 has been discontinued as part of this release. This change helps:

• • Ensure the platform remains supported and future-ready.
  • Align with the lifecycle of the underlying SDK.
  • Provide a more stable and forward-compatible environment for modern workloads.

• A More Powerful and Up-to-Date Ecosystem

One of the biggest highlights of this release is the comprehensive update of core packages across the AI and data science stack.

What this means for you:

• • Better performance for data processing and model training.
  • Updated dependencies.
  • Greater compatibility with modern libraries and frameworks.
  • Reduced friction when porting workloads to z/OS.

• Expanded Capabilities with New Packages

To support newer use cases and developer productivity, several new packages have been introduced:

• • python-dotenv – Simplifies environment configuration management
  • rapidfuzz – High-performance fuzzy string matching
  • scikit-base – Foundational utilities for ML frameworks
  • sktime – Time series analysis and forecasting
  • starlette – Lightweight ASGI framework for building APIs
  • wheel – Enhanced packaging and distribution support

These additions make it easier to build modern applications ranging from APIs to advanced ML pipelines—directly on z/OS.

• Streamlined for Reliability

As part of ongoing efforts to maintain a reliable toolkit, the markdown-editor package has been removed in this release, as previously communicated.

Getting Started

Getting started with Python AI Toolkit 1.14 is simple:

Use the requirements.txt file provided in repository

• Select the file corresponding to your Python version.
• Install and start building with the latest toolkit capabilities.

Additionally, the official documentation has been updated with the latest package information.

Why Upgrade to 1.14?

Whether you're building new solutions or modernizing existing ones, this release helps you:

• Stay current with the latest Python ecosystem
• Accelerate AI/ML workloads on enterprise systems
• Improve security and stability
• Leverage a richer and more flexible open-source ecosystem

Looking Ahead

We continue to focus on:

• Keeping pace with open-source innovation.
• Enhancing developer experience on z/OS.
• Expanding AI and ML capabilities for enterprise workloads.