Google Agent Development Kit
adk.dev
- Free plan
- Open source
- Windows
- macOS
- Linux
- Web
- API
- Self-hosted
What Google Agent Development Kit is
Google Agent Development Kit (ADK) is an open-source framework for building, debugging, evaluating, and deploying AI agents. It supports Python, TypeScript, Go, Java, and Kotlin, with workflows that combine fixed code paths and adaptive AI reasoning. Teams can arrange agents in hierarchies so they delegate tasks to specialists. ADK works with Gemini and adapters for other model providers, including locally running models, and its integrations cover MCP, Google Cloud, databases, search, observability, and third-party services. Evaluation tools let teams build multi-turn datasets and run checks locally from the CLI or development UI. For live agents, it supports two-way text and audio streaming, and it manages session context with links to longer-term memory services. Deployment options include Google Cloud Agent Runtime, Cloud Run, GKE, and container-friendly environments such as local Docker or Podman. ADK is free; model and cloud service use may require separate access or billing. Its security guidance covers identity, authorization, guardrails, sandboxed execution, and network controls. Agent Config is narrower: it currently supports Gemini models only and requires Python.
Who it suits
ADK suits developers building AI agents who want to work in one of its five supported languages and choose among cloud, container, or local deployment. Teams needing agent hierarchies, evaluation tools, or live voice interactions may also find its capabilities relevant.
Good to know, and what to watch
- Supports five programming languages.
- Agents can delegate to specialized agents.
- Local evaluation through CLI or development UI.
- Supports live two-way text and audio.
- Can deploy to cloud or local containers.
- Agent Config supports Gemini models only.
- Agent Config requires Python.
- Model and cloud usage may require separate billing.
Plans
Every plan and priceADK framework
Google Agent Development Kit (ADK)
Free plan
The verdict
ADK brings agent building, evaluation, and deployment into an open-source framework with multiple language and hosting options. Check the Agent Config restrictions and any separate model or cloud requirements before choosing it.
What the maker publishes
- Agent orchestration
- The framework supports modular multi-agent composition, delegation and routing between specialized agents. source
- API surface
- Official references include Python, TypeScript, Go, Java and Kotlin APIs plus CLI, configuration and REST API references. source
- CLI
- Agents CLI can scaffold, build, test, evaluate and deploy ADK agents from an AI-enabled development environment. source
- Company founded
- Google was founded in 1998 by Larry Page and Sergey Brin. source
- Company legal identity
- Google LLC is Alphabet's technology subsidiary and is headquartered in Mountain View, California. source
- Company size
- LinkedIn lists Google as a public software-development company with 10,001-plus employees and headquarters in Mountain View, California. source
- Context and memory
- ADK manages session context and provides integration points for longer-term memory services. source
- Context management
- ADK manages sessions, memory, events, tool outputs and artifacts, including summarization and context compression controls. source
- Deployment
- Agents can be deployed to Google Cloud Agent Runtime, Cloud Run, GKE, or other container-friendly infrastructure, including local Docker or Podman environments. source
- Deployment
- Agents can run locally, in containers, on Cloud Run, on GKE or on Google's managed Agent Runtime. source
- Developer UI
- The local ADK web interface lets developers chat with agents and inspect events, state and execution traces step by step. source
- Documentation depth
- The documentation site includes getting-started guides, component references, deployment, observability, evaluation, safety, MCP, A2A and API references. source
- Evaluation
- ADK includes tools to create multi-turn evaluation datasets and run evaluations locally through the CLI or development UI. source
- Evaluation
- Documentation covers evaluation criteria, user and environment simulation, custom metrics and optimization workflows. source
- Graph workflows
- ADK 2.0 adds graph-based workflows that combine deterministic code with model reasoning and explicit execution paths. source
- Installation
- The installation guide provides package installation instructions for all five supported languages. source
- Installation requirement
- The Python quickstart requires Python 3.10 or later and installs the package with pip install google-adk. source
- Integration ecosystem
- Google lists integrations spanning Google Cloud services, databases, observability platforms, communication tools, productivity apps and many third-party agent tools. source
- Integrations
- The integrations catalog includes tools and connectors for MCP, Google Cloud, databases, search, observability, and third-party services. source
- Languages
- ADK is available for Python, TypeScript, Go, Java, and Kotlin. source
- Languages
- Official implementations are provided for Python, TypeScript, Go, Java and Kotlin. source
- Launch news
- Google announced ADK on April 9, 2025 at Google Cloud NEXT as an open-source framework for production agent and multi-agent development. source
- License
- The Python ADK repository is licensed under the permissive Apache License 2.0. source
- Limit: beta components
- ChatForest reports that Python 2.0 and Kotlin 0.1.0 were beta releases and that ADK for Android was not production-ready at review time. source
Checked against the maker's pages on 15 Oct 2026.