# Google ADK Python 2.14.0 graduates environments, A2A and eval; 23% faster workflows

Canonical: https://theagenticweekly.com/news/google-adk-python-2-14-0 · Evidence: Reported · First seen: 2026-11-07 01:04 UTC · Source published: 2026-11-07 00:58 UTC · Brief published: 2026-11-07 07:16 UTC · Updated: 2026-11-07 07:16 UTC · Author: Agentic Weekly

> Google released ADK Python 2.14.0. The release notes say the cut "graduates execution environments, A2A, evaluation, and core runtime modules from experimental" and that a 30-node workflow's 200-turn run is "23% faster with 9% lower peak memory," with TeamsRunner, OpenAI Decisions, MultiRunConfig and ToolArgsMatcher added.

## What changed

- Stable APIs: the release notes say 26 auth, environment, app, CLI and cache modules plus 16 toolset and integration features, and mature evaluation and user simulation classes, graduate from experimental.
- Performance: the notes report a 30-node workflow's 200-turn run is 23% faster with 9% lower peak memory after long multi-turn workflow work.
- Evaluation: MultiRunConfig (adk eval --num_runs) and --parallelism run each eval case multiple times; ToolArgsMatcher compares tool-call arguments with exact, numeric-tolerance or LLM-backed semantic checks.
- Integrations: TeamsRunner connects an ADK agent to Microsoft Teams chats and channels; OpenAIDecisionsClassifier uses the OpenAI Decisions API for workflow and tool-gating questions; A2A is stable with adk generate_agent_card.
- Breaking: FunctionTool parameters annotated Enum, Optional[Enum] or list[Enum] now receive enum members instead of raw strings; VertexAiSessionService.create_session only accepts listed session config options; RunSkillScriptTool in environment mode no longer takes a free-form command.

## In the source's words

> "This release graduates execution environments, A2A, evaluation, and core runtime modules from experimental, adds multi-run evaluation and semantic tool argument matching, adds Microsoft Teams and OpenAI Decisions integrations, and makes long multi-turn workflows faster."
> (Google ADK Python GitHub release v2.14.0, https://github.com/google/adk-python/releases/tag/v2.14.0)

> "a 30-node workflow's 200-turn run is 23% faster with 9% lower peak memory"
> (Google ADK Python GitHub release v2.14.0 (Highlights), https://github.com/google/adk-python/releases/tag/v2.14.0)

> "26 auth, environment, app, CLI and cache modules, 16 toolset and integration features, and the mature evaluation and user simulation classes graduate from experimental"
> (Google ADK Python GitHub release v2.14.0 (Stable APIs), https://github.com/google/adk-python/releases/tag/v2.14.0)

## What we have not verified

- We did not install google-adk 2.14.0 or re-run the 30-node / 200-turn workflow benchmark ourselves.
- We did not verify TeamsRunner, OpenAI Decisions, A2A card generation or ToolArgsMatcher beyond the release notes.
- Migration cost for enum-as-members and RunSkillScriptTool command removal was not measured.

## Timeline

- Source published: 2026-11-07 00:58 UTC
- First seen by our watcher: 2026-11-07 01:04 UTC (via GitHub releases: google/adk-python)
- Brief published: 2026-11-07 07:16 UTC

## Sources

- [GitHub release: google/adk-python v2.14.0](https://github.com/google/adk-python/releases/tag/v2.14.0) (primary source) (fetched 2026-11-07)

## Related

- [Google ADK: verdict and setup guide](https://theagenticweekly.com/tools/google-adk)
- [Semantic Kernel vs Google ADK](https://theagenticweekly.com/compare/semantic-kernel-vs-google-adk)
- [Alternatives to Google ADK](https://theagenticweekly.com/alternatives/google-adk)
- [All news briefs](https://theagenticweekly.com/news)
- [This week's releases](https://theagenticweekly.com/this-week)
- [New AI models this week](https://theagenticweekly.com/models/this-week)

## How to cite

Agentic Weekly, "Google ADK Python 2.14.0 graduates environments, A2A and eval; 23% faster workflows", published 2026-11-07 07:16 UTC, https://theagenticweekly.com/news/google-adk-python-2-14-0