Yes, RAGFlow includes a visual workflow builder that enables low-code, no-code design of RAG and agentic pipelines. From v0.8 onward, RAGFlow officially entered the Agentic era with a comprehensive graph-based task orchestration framework on the backend and a no-code workflow editor on the frontend. The visual editor lets you drag components onto a canvas—document ingestion, chunking, knowledge graph construction, embedding, retrieval, LLM calls, agents—connect them with data flow lines, and configure parameters through UI forms. This approach makes RAGFlow accessible to non-developers and dramatically reduces configuration time compared to writing orchestration code. The workflow builder understands RAGFlow’s native components (semantic chunking, hybrid search, re-ranking, knowledge graphs) and exposes their key parameters visually. Complex workflows like multi-stage retrieval, agent feedback loops, and conditional logic can be composed without coding. The visual paradigm encourages experimentation—you can quickly iterate on pipeline designs, test different chunking strategies, adjust re-ranking weights, and see impacts through the UI. Advanced users can still access APIs for programmatic workflows, but the visual builder is the preferred path for most users. This no-code capability is one of RAGFlow’s key differentiators from code-first frameworks like LangChain, making RAGFlow ideal for teams prioritizing speed and accessibility over ultimate customization.
Related Resources: RAG Pipeline with Milvus | Improving Chunking for RAG