open-design-v0.22.1 v2.3.0

medium risk

data-privacy-stack/presidioSep 9, 2026 → Sep 9, 20260 days between releases

What's New

v0.22.1 enables OD Next defaults; v2.3.0 adds BigQuery and Vertex AI support.

Breaking Changes

  • Google models now use Vertex AI, requiring migration for deployments using Google AI Studio credentials.

New Features

  • OD Next is now the default for supported design tasks.
  • BigQuery federation supports temporal expressions and window aggregates.
  • Caching accelerator now has explicit size and item limits.
  • SQL results cache supports stale-while-revalidate across refreshes.
  • GitHub connector adds tables and columns for review, release, and repository data.

Improvements

  • OD Next provides more concise tool guidance and clearer completion instructions.
  • Failed runs now retain timeout and retry information with Retry actions.
  • Improved error messaging for API keys and unavailable models.
  • Enhanced project references and Docker plugin browsing capabilities.
  • SQL federation improves function translation, NULL handling, and timezone declarations.

Bug Fixes

  • Failed runs now retain timeout and retry information.
  • Fixed function translation, NULL handling, and timezone declarations in SQL federation.
  • Failed media work no longer displays a completed checkmark incorrectly.

Upgrade Notes

If using Google models, migrate credentials from Google AI Studio to Vertex AI service accounts. Configure new caching limits `caching_max_size` and `caching_max_items` for accelerators.

The Two Releases

open-design-v0.22.1

Open Design 0.22.1 — Smoother design runs and recovery

Sep 9, 2026· 1 month ago

Builds on the previous version by enabling OD Next by default for supported design tasks and significantly improving recovery mechanisms for failed runs and error messaging.

Full release notes
v2.3.0

Spice v2.3.0 (September 9, 2026)

Sep 9, 2026· 1 month ago

This release introduces broader SQL federation capabilities, particularly extending BigQuery support and adding a GitHub connector. It also adds explicit size and item limits to caching accelerators and improves cache handling to support stale data across refreshes.

Full release notes