ragent and java-rag
About ragent
nageoffer/ragent
RAG综合智能体 - 基于Spring Boot的智能文档处理与检索系统，集成向量数据库，拥有智能问答、知识库、会话记忆、深度思考等功能
Implements a production-grade RAG pipeline with parallel multi-channel retrieval (semantic + keyword-based), model routing across multiple LLM providers with automatic failover, and document ingestion via composable node-based ETL. Integrates Milvus for vector search, RocketMQ for async processing, and supports MCP tool invocation alongside knowledge retrieval, with full request tracing across intent recognition, query rewriting, and response generation stages.
About java-rag
ChinaYiqun/java-rag
This RAG (Retrieval-Augmented Generation) project is implemented using pure Java. This approach makes it easier to adapt to enterprise-level environments and is more conducive to secondary development.
Implements modular RAG pipelines with pluggable components for document parsing (PDF, Word, PPT, Excel), multiple chunking strategies (fixed-size, semantic, recursive), and vector search across Elasticsearch, Redis, and MinIO. Supports both OpenAI and Ollama LLM interfaces with multi-turn conversation management, load balancing, and Agent patterns—all configured via Nacos for enterprise flexibility without framework dependencies.
Related comparisons
Scores updated daily from GitHub, PyPI, and npm data. How scores work