Reported News
Google ADK Python 2.14.0 graduates environments, A2A and eval; 23% faster workflows
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.
- Changes reported
- 5
- listed below, each from a source we read
- We tested
- Not tested yet
- reported from sources; we did not run it ourselves
- Gaps we flag
- 3
- things we could not verify
- Sources
- 1
- 1 primary, linked below
What changed
| # | What the source says | Status |
|---|---|---|
| 01 | 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. | Reported |
| 02 | 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. | Reported |
| 03 | 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. | Reported |
| 04 | 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. | Reported |
| 05 | 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. | Reported |
reported by the sourcemeasured or run by usnot verified
Primary source: GitHub release: google/adk-python v2.14.0. We are the fast, independent index; the vendor’s own page is the source of record for this release.
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.”
“a 30-node workflow's 200-turn run is 23% faster with 9% lower peak memory”
“26 auth, environment, app, CLI and cache modules, 16 toolset and integration features, and the mature evaluation and user simulation classes graduate from experimental”
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.
This brief is Reported: written from the sources below, not from running the release ourselves.
Sources
- GitHub release: google/adk-python v2.14.0primary source · fetched 2026-11-07
Related
Spotted an error? See our corrections policy or email the address on the contact page. Markdown version for agents: /news/google-adk-python-2-14-0.md
Want these briefs in your Google Top Stories? Add Agentic Weekly as a preferred source on Google.
Newsletter
The Weekly Verdict. Free, every Monday.
A free email every Monday: the week's few big AI releases, then 5 short use cases for the new models, each with a prompt you can paste.
Free. Double opt-in. One click to unsubscribe.