Return to page


What's new in the latest release of H2O AI Hybrid Cloud?


By Michelle Tanco | minute read | April 25, 2023

Blog decorative banner image

Check out the complete release notes here! v23.01.0 | Apr 14, 2023 

Upgraded Components

Core Components

AI App Store v0.22.0 AI App Store v0.22.0

The AI App Store is a platform for accessing and operationalizing AI/ML applications and services that are built using H2O Wave .
The 23.01.0 Hybrid Cloud release introduces multiple UI enhancements to make the user experience more efficient and intuitive.

  • ‘My Apps’ page
    • Added support for Python Apps
    • Easily manage apps from the UI
    • Find the app you’re looking for more easily
    • Easily copy the UUID
  • ‘My Instances’ page
    • Find instances more easily
    • Quickly find all versions of the exact app you’re seeking
    • Enhanced app filtering
  • ‘Admin Apps’ page
    • Find the app you’re looking for more easily
    • Easily copy the UUID
  • ‘Admin Instances’ page
    • Sort app instances in the admin instances UI
    • Improved scalability of app instances
H2O Driverless AI v1.10.4.2 and v1.10.4.3 H2O Driverless AI v1.10.4.2 and v1.10.4.3
  • DAI consistently delivers industry-leading AutoML capabilities specifically designed to use AI to make AI. Driverless AI can be configured to run in a multinode worker mode (see multinode userguide here).
    • v1.10.4.2 provides the ability to use self-signed certificates or skip the certificate validation for OPENID authentication.
    • v1.10.4.3 allows users to generate and use MapR ticket with different identities and delivers multiple bug fixes.
  • H2O-3 v3.38.0.3 and v3.38.0.4
H2O-3 v3.38.0.3 and v3.38.0.4 H2O-3 v3.38.0.3 and v3.38.0.4

H2O-3 is an open-source, in-memory, distributed, fast, and scalable machine learning and predictive analytics platform allowing you to build big data machine learning models.

  • Key updates that come with the addition of v3.38.0.3 and v3.38.0.4
    • Implemented p-value calculation for GLM with regularization
    • Implemented normal (non-monotonic) splines that can support any degrees
    • Verified the minimum number of knots each spline type can support for GAM
  • H2O MLOpsv0.59 and 0.60.1
H2O MLOps v0.59 and 0.60.1 H2O MLOps v0.59 and 0.60.1

H2O MLOps provides a collaborative environment for Data Scientists and IT teams to manage, deploy, govern, and monitor machine learning models.
MLOps v0.59  and v0.60.1  shipped with the Hybrid Cloud 23.01.0 release.

  • v0.59 Highlights

    • Deployment-related improvements introduced beginning with v0.59.0
    • Many configuration options that were previously static are now configurable.
    • Deployed scoring applications now set additional Kubernetes annotations.
    • Users can download Kubernetes logs from deployments in the MLOps Wave App and MLOps API.
  • Storage and Scoring Telemetry
    • MLOps can now send analytical data related to storage operations to the telemetry server.
    • MLOps Scoring now sends scoring-related data to the telemetry server.
    • Static scoring endpoints: You can now define and update a persistent URL pointing to a particular MLOps deployment.
  • v0.60.1 Highlights

    • Introduction of a feature flag that enables the import of third-party experiments (pickled experiments) flow with Conda.
    • Enhancements
      • Search for users by username when sharing a project with another user. Sort the user list in alphabetical order.
      • Model monitoring – feature summary table: figures are now displayed only up to three decimals places.
      • The deployment ID is conditionally displayed when no deployment name is present.
      • When keycloak is unavailable, a blocking error page is now displayed.
Enterprise Steam v0.1.9.0 Enterprise Steam v0.1.9.0
  • Enterprise Steam enables streamlined adoption of products in a secure manner that complies with company policies.
    • v1.9.0 Highlights
      • Driverless AI: Added support for single-user multinode clusters and fixed maximum uptime detection
      • Python: Removed tokens from Python client logging

Optional Components

These components are optional additions to the H2O AI Cloud – reach out to your H2O representative or for more information. 

  • H2O Drive/H2O Exchange Store v0.7.2 (currently only available as a preview for Hybrid Cloud users on AWS)
H2O Drive/H2O Exchange Store v0.7.2 (currently only available as a preview for Hybrid Cloud users on AWS) H2O Drive/H2O Exchange Store v0.7.2 (currently only available as a preview for Hybrid Cloud users on AWS)

H2O Drive is a secure personal object-store for H2O AI Cloud.
Data in H2O Drive can be used within many apps, meaning you only have to bring data into the H2O AI Cloud once.

  • v0.7.2 Highlights
    • Users can now upload *.ipynb Jupyter Notebook files using the local file upload connector.
    • Convert http requests to https when using the Amazon S3 connectors
  • H2O Feature Store v0.14.4 (currently only available as a preview for Hybrid Cloud users on AWS and Azure)
H2O Feature Store v0.14.4 (currently only available as a preview for Hybrid Cloud users on AWS and Azure) H2O Feature Store v0.14.4 (currently only available as a preview for Hybrid Cloud users on AWS and Azure)
  • The H2O Feature Store allows engineers to streamline data quality management across all machine learning pipelines, reducing the time spent on repetitive tasks.
    • v0.14.4 Highlights
      • JWT token no longer requires an expiration date to ensure a consistent user experience
      • Sensitive consumer permissions are not granted if a user is a regular consumer
  • H2O Document AI v0.5.0
H2O Document AI v0.5.0 H2O Document AI v0.5.0

Document AI combines several machine learning disciplines to include optical character recognition, natural language processing, handwritten text recognition, text extraction, and more to validate data, analyze documents and help humans extract information from those documents meaningfully.

  • v0.5.0
    • New Features
      • Optical character recognition (OCR) language support for
        • Latin (e.g., Spanish)
        • Arabic (e.g., Persian)
      • Document Text Recognition (DocTR) EfficientNet models to better recognize handwritten documents
      • Enhancements
        • Upgraded the ML API to v0.4.0
        • Refactored and improved the training user interface for better usability
        • Ability to gate access to H2O Document AI based on a user’s role
  • H2O Hydrogen Torch v1.2.0
H2O Hydrogen Torch v1.2.0 H2O Hydrogen Torch v1.2.0

H2O Hydrogen Torch is an application that lets novice and expert data scientists build deep learning models for diverse problem types  in computer vision, natural language, and audio. No code is required.

  • v0.1.2.0 Highlight Summary
    • UI & UX – enhanced user experience
    • Datasets
      • Data connector: Hydrogen Torch now supports Azure data lake (as a data connector).
      • Image object detection: Hydrogen Torch now supports several dataset (data) formats for an image object detection experiment.
      • And much more…
    • Experiments
      • H2O Hydrogen Torch offers several grid search modes
      • Users can enter custom values for any grid search hyperparameter values
      • And much more…
    • *Please visit the release notes for complete details

New Components

AI Engine Manager (AIEM)  is an exciting new optional component and is recommended for new installations of Hybrid Cloud.

AI Engine Manager (AIEM) AI Engine Manager (AIEM)

The goal of AIEM is to be the new go-to central service integrating with Kubernetes for model-building tools like H2O Driverless AI and H2O-3, enabling other teams to focus on making the best ML products instead of managing Kubernetes resources.
Unlike H2O Enterprise Steam, AIEM features an interface found directly within the Hybrid Cloud interface to improve user flow.

Note:  If you are currently using or want to use an older version of H2O Driverless AI or H2O-3, using Enterprise Steam instead of AIEM is recommended. 


Michelle Tanco, Head of Product

As the Head of Product at, Michelle Tanco’s primary focus lies in delivering a seamless user experience across machine learning applications. With a strong dedication to math and computer science, she is enthusiastic about leveraging these disciplines to address real-world challenges. Before joining H2O, she served as a Senior Data Science Consultant at Teradata, working on high-impact analytics projects to tackle business issues across various industries. Michelle holds a B.A. in Mathematics and Computer Science from Ursinus College. In her downtime, she enjoys spending quality time with her family and expressing her creativity by playing the bass and ukulele.