Google released Agent Development Kit 2.14.0 on November 7 — the release notes are dated November 4 and the tag was published November 7. Execution environments, the A2A agent-to-agent protocol, evaluation and the core runtime modules graduate from experimental to stable, alongside 26 auth, environment, app, CLI and cache modules and 16 toolset and integration features.

The release adds a TeamsRunner that connects ADK agents to Microsoft Teams chats and channels, a MongoDbMemoryService with vector-search retrieval and TTL expiration for agent memory, multi-run evaluation, and an adk generate_agent_card command that builds A2A agent cards from the CLI. A classifier for workflow and tool-gating questions uses OpenAI's Decisions API.

Google claims a 30-node workflow running 200 turns is 23% faster with 9% lower peak memory in this release. The release notes publish no baseline hardware, no comparison version and no second benchmark, so treat that figure as Google's own unverified claim. Stable status sets an API baseline for agent builders; it is not by itself a guarantee that an application built on it is production-ready.