- Published on
Embabel Agent Framework Reaches 1.0
- Authors
- Name
- DeepHorus
Embabel, a Java framework for building AI agents, reached its 1.0 general-availability release, marking a milestone for Java and Kotlin developers. Co-created by Rod Johnson, creator of the Spring Framework, Embabel lets developers define agents as typed domain objects with goals, actions, and conditions, rather than hand-coding sequences of prompts and tool calls. This release provides a new way to structure intelligent workflows in the JVM ecosystem.
The framework borrows Goal-Oriented Action Planning (GOAP) from video game AI. Instead of following a pre-wired graph, the agent gets a set of actions with preconditions and effects, and a planner searches for a sequence satisfying the goal. If the world changes mid-task—such as a tool call failing—the planner reassesses and finds a new path without breaking.
Embabel is built on top of Spring AI, not replacing it. Johnson's README draws a parallel: Spring AI is like the Servlet API, while Embabel is like Spring MVC. It inherits support for OpenAI, Anthropic, Gemini, Bedrock, Mistral, DeepSeek, and local options via Ollama, Docker, or LMStudio. Developers can pin individual actions to specific models or use role aliases for cost and capability routing.
This approach sets Embabel apart from LangGraph, which models workflows as directed graphs defined up front. Java teams can use LangGraph4j, but Embabel's runtime planner can combine actions into sequences developers never explicitly wired. Embabel also supports mixing GOAP planning with explicit state machines.
Johnson announced the release with excitement, and the project's history draws direct parallels to Spring's early days. The framework aims to solve recurring agent problems, just as Spring MVC solved HTTP handling. Now at 1.0, Embabel offers a production-ready alternative for Java teams building complex, adaptive agents.
With multi-model routing and dynamic planning, Embabel is positioned to become a key tool in the Java AI ecosystem. Its success will depend on community adoption and how well it complements Spring AI in real-world deployments. For developers, it promises a higher-level, more resilient approach to agent orchestration.
