Skip to content
AI.info

tools

H2O Driverless AI

H2O Driverless AI automates machine learning workflows, including feature engineering, model building, explainability, and deployment.

H2O Driverless AI

H2O Driverless AI is an automated machine learning platform for data scientists, IT teams, and business analysts. It imports data, transforms features, trains and validates models, explains predictions, and supports deployment.

It can export scoring pipelines as Python modules or Java artifacts and expose models through REST endpoints or cloud services. The product requires a commercial license; H2O.ai does not publish public tier pricing on the pages reviewed.

Features

  • Automates feature engineering and feature selection
  • Builds, tunes, validates, and selects machine learning models
  • Provides global and local model explanations
  • Includes automatic data visualization and quality checks
  • Supports custom recipes and time-series modeling
  • Exports Python and Java scoring pipelines
  • Deploys models through REST endpoints, cloud services, or edge artifacts
  • Provides a Python client and API

Use cases

  • Build predictive models from tabular business data
  • Detect fraud and other unusual transactions
  • Forecast demand, revenue, or operational metrics
  • Explain model predictions for regulated decisions
  • Deploy scoring pipelines in applications or edge systems
  • Help analysts create models with guided workflows

Pros

    Cons

      Latest updates

      • Version 2.5.1 (September 29, 2026) (2.5.1)

        Added early stopping to CatBoost fitting; increased image fine-tuning epochs; fixed target scaling, transformer fine-tuning, imports, and date sorting.

      • Version 2.5.0 (August 28, 2026) (2.5.0)

        Added DAICatBoost as a built-in model, batch TTA support, and automatic seasonality and trend detection.

      • Version 2.4.5 (July 29, 2026) (2.4.5)

        Updated the custom H2O-3 build and HMLI to Java 17 / Jetty 12; upgraded MOJO2 runtime and transformers library to 5.x.

      • Version 2.4.4 (June 27, 2026) (2.4.4)

        Migrated pretrained-model download URLs from S3 to CDN and upgraded PyTorch Lightning and AutoDoc.

      • Version 2.4.3 (May 26, 2026) (2.4.3)

        Raise error message in UI when test dataset used by Model Diagnostic does not contain target column.

      Capabilities

      • Charts and dashboards — “AutoViz Automatic visualization of the dataset before modeling: distributions, outliers, correlations, and missing values.” source
      • Connects to your data — “01 Connect data Files, JDBC, Snowflake, Databricks, S3, or Hadoop.” source
      • Builds models — “The leading AutoML platform that automates feature engineering, model selection, and tuning, and documents every model as it trains.” source
      • Real-time data — “Python pipelines for batch and low-latency Java MOJO for real time, deployable to MLOps, Hadoop, Spark, Teradata, or Oracle” source
      • Self-hosted — “Deployment On-premises, air-gapped, private cloud, or cloud.” source

      Get it

      Security

      • SOC 2 Type II — “SOC 2 Type 2” source
      • SOC 2 — “SOC 2 Type 2” source
      • ISO 27001 — “ISO/IEC 27001:2022” source
      • GDPR — “Data Processing Agreement” source

      Pricing

      Prices checked
      2026-10-24
      Official website