opensolon/solon-ai

Java AI application development framework (supports LLM-tool,skill; RAG; MCP; Agent-ReAct,Team-Agent). Compatible with java8 ~ java26. It can also be embedded in SpringBoot, jFinal, Vert.x, Quarkus, and other frameworks.

What it solves

Solon AI is a comprehensive Java-native AI development framework designed to bring the capabilities of Python-based frameworks like LangChain or LlamaIndex to the Java ecosystem. It allows Java developers to build production-grade AI applications—such as autonomous agents, RAG knowledge bases, and multi-agent systems—without leaving their preferred language and environment.

How it works

The framework provides a unified interface to interact with various LLM providers (OpenAI, Gemini, Claude, Ollama, etc.) through a "dialect" system that handles model differences. It organizes AI logic into several core modules:

  • ChatModel: A general-purpose interface for LLM calls, supporting synchronous and reactive streams.
  • RAG Pipeline: A full-link system including document loaders, splitters, embedding models, and reranking models for knowledge retrieval.
  • MCP Integration: Deep support for the Model Context Protocol, allowing the creation of both MCP servers and clients for cross-platform tool and resource sharing.
  • Agent Orchestration: Implements reasoning patterns like ReAct (Think-Call-Observe-Summarize) and Team collaboration via predefined protocols (e.g., hierarchical structures).
  • Ai Flow: A YAML-based orchestration engine that enables low-code, graph-driven workflow definitions similar to Dify.

Who it’s for

Java developers and enterprise teams who need to integrate LLMs and agentic workflows into existing Java applications (e.g., SpringBoot, Vert.X, Quarkus) across a wide range of JDK versions (8 through 25).

Highlights

  • Java-Native Ecosystem: Seamlessly integrates with major Java frameworks and supports a broad spectrum of JDK versions.
  • Multi-Model Dialects: Unified API for multiple LLM providers, reducing vendor lock-in.
  • Graph-Driven Agents: Transforms complex agent reasoning into observable and governable computation flow graphs.
  • MCP Support: Native implementation of the Model Context Protocol for standardized tool sharing.
  • Low-Code Flow: YAML-based process orchestration for rapid AI workflow prototyping.

Related

  • Project
    agents-flex/agents-flex

    A lightweight, modular AI application framework for Java that provides unified abstractions for LLMs, RAG, and agents, with production-ready tooling for Spring Boot integration.

  • Project
    LnYo-Cly/ai4j

    A Java AI Agentic SDK for JDK 8+ that provides unified access to multiple LLM platforms, tool calling, RAG, and agent runtime capabilities.

  • Project
    JetBrains/koog

    A Kotlin-based framework for building and running AI agents with multiplatform support, graph-based workflows, and seamless LLM provider switching.

  • Project
    ThomasVitale/llm-apps-java-spring-ai

    A collection of sample Java applications and patterns demonstrating how to build Generative AI and LLM-powered apps using the Spring AI framework.

  • Project
    cloudwego/eino

    An LLM application development framework for Golang that provides tools for building AI agents, multi-agent coordination, and graph-based workflows.