This project provides an agentic AI framework for Java developers, specifically integrating with Alibaba Cloud's QWen LLM services and cloud-native infrastructures. It simplifies adding generative AI capabilities to Spring Boot applications, targeting Java developers looking to build AI-powered applications.

How It Works

The framework leverages Spring Boot 3.x and provides high-level abstractions like ChatClient for seamless interaction with various AI models, including Alibaba Cloud's QWen and Dashscope services. It supports multiple model types (Chat, Text-to-Image, Audio Transcription, Text-to-Speech) with both synchronous and streaming APIs. Key features include function calling, POJO mapping for model outputs, portable APIs for Vector Stores, and RAG support with components for document processing, embedding, and retrieval.

Quick Start & Requirements

• Install: Add com.alibaba.cloud.ai:spring-ai-alibaba-starter dependency to your Maven project.
• Prerequisites: JDK 17+, Maven. Requires adding the https://repo.spring.io/milestone repository.
• Setup: Minimal setup involves adding the dependency and repository.
• Docs: Quick start

Highlighted Details

• Supports Alibaba Cloud QWen and Dashscope models.
• Offers RAG support with document processing, embedding, and vector store integration.
• Provides function calling and conversation memory for agentic behavior.
• Includes Spring Boot auto-configuration and starters for easy integration.

Maintenance & Community

• Community-driven project with a focus on Java implementations.
• References Spring AI and Langgraph4J for inspiration.
• Contact via Dingtalk Group (61290041831) and WeChat.

Licensing & Compatibility

• Licensed under Apache License 2.0.
• Compatible with commercial use and closed-source linking.

Limitations & Caveats

The project is currently in milestone release (1.0.0-M6.1), indicating potential for breaking changes. Spring AI related packages are not yet published to the central Maven repository, requiring manual repository configuration.