Google · 2026-04-13 · notable
Google ADK v1.30 — Auth Providers, Gemma 4, and BigQuery Tools GA
Google's Agent Development Kit ships auth provider support for agent registries, Gemma 4 model integration, live avatar features, and promotes BigQuery tools to stable — plus a security fix for credential leakage.
Google's agent framework adds auth provider support and Gemma 4 models while graduating BigQuery tools to production-ready.
Key specs
| Version | 1.30.0 |
|---|---|
| GitHub stars | 19k |
| Languages | Python, TypeScript, Go, Java |
What is it?
Agent Development Kit (ADK) is Google's open-source, code-first framework for building AI agents. Version 1.30.0 adds authentication provider support to agent registries, native Gemma 4 model integration, artifact service access in agent-to-agent events, and live avatar support. BigQuery tools move from experimental to stable.
How does it work?
ADK lets you define agent logic, tools, and orchestration directly in Python (or TypeScript, Go, Java) with modular multi-agent hierarchies. The new auth provider integration allows agents to authenticate with external services through the agent registry. Session resumption is now exposed as an event in the base LLM flow, making it easier to build persistent agents. The release also fixes a credential leakage vulnerability and adds path traversal validation.
Why does it matter?
ADK has become Google's answer to LangChain and similar frameworks, but with tighter Vertex AI and Gemini integration. The auth provider feature addresses a real deployment gap — most agent frameworks punt on authentication. BigQuery tools graduating to stable is a signal that enterprise data workflows are a first-class use case.
Who is it for?
Developers building multi-agent systems on Google Cloud or with Gemini models.
Try it
pip install google-adk==1.30.0