Google releases agent development kits for Python and Java

Google has released agent development kits for the Python and Java programming languages.

The Python ADK 1.0.0 and Java ADK 0.1.0 were announced on May 20. The open source tool kits are designed to be used for building and deploying sophisticated AI agents with flexibility and control, Google said. For Python, Google said the milestone of the 1.0.0 release signifies that the Python ADK is now production-ready, offering a reliable and robust platform for developers to confidently build and deploy their agents in live environments. For Java, Google said development of the kit brings the power and flexibility of the ADK to Java developers, enabling them to leverage its capabilities for agent development needs.

Google describes the ADK as a flexible and modular framework for developing and deploying AI agents. While optimized for Gemini and the Google ecosystem, ADK is model-agnostic and deployment-agnostic and built for compatibility with other frameworks, Google said. The ADK was designed to make agent development feel more like software development, to make it easier for developers to create, deploy, and orchestrate agentic architectures that range from simple tasks to complex workflow.

Key features of the Python and Java ADKs include the following:

A rich tool ecosystem, with developers able to use prebuilt tools, custom functions, and OpenAPI specs, or integrate existing tools to give agents diverse capabilities, all for tight integration with the Google ecosystem.

Code-first development, with developers able to define agent logic and orchestration directly in Java or Python for flexibility, testability, and versioning.

Modular multi-agent systems, to define scalable applications by composing multiple specialized agents into flexible hierarchies.