ageerle/ruoyi-ai

面向企业级市场的一站式AI应用开发框架，支持多厂商大模型统一接入与管理，具备安全可控的企业知识库与高精度检索优化能力，提供可视化流程编排、自主决策智能体与多智能体协同调度，兼容主流 Agent Skill 协议，帮助企业与开发者零门槛快速构建安全、高效、可落地的AI智能体应用与行业解决方案。

What it solves

RuoYi AI provides an out-of-the-box, enterprise-grade full-stack AI platform that simplifies the deployment of AI assistants. It integrates multiple LLM providers, knowledge bases, and agent orchestration, removing the need to build these components from scratch for business applications.

How it works

The platform is built on a Spring Boot 3.5.8 backend and a Vue 3 frontend. It uses Langchain4j as its core agent framework to implement a Supervisor mode for agent orchestration and various decision-making models. For knowledge management, it employs a RAG (Retrieval-Augmented Generation) system using vector databases like Milvus, Weaviate, or Qdrant to handle document parsing and retrieval.

Who it’s for

Developers and enterprises looking to quickly deploy a full-stack AI assistant platform with integrated management tools for models, knowledge, and workflows.

Highlights

  • Multi-Model Support: Integrates with DeepSeek, Zhipu, MIMO, Bailian, and OpenAI, as well as platforms like Coze, DIFY, and FastGPT.
  • ** لقب Supervisor Mode**: Supports multi-agent collaboration and complex orchestration via a Supervisor pattern.
  • Visual Workflow Designer: A drag-and-drop interface for designing workflows with nodes for model calls, email sending, and human review.
  • Comprehensive RAG: Local RAG capabilities with support for multiple vector databases and parsing for PDF, Word, and Excel files.
  • Extensible Tooling: Integration with the MCP protocol and a customizable skills ecosystem.