Skip to content

语析Yuxi-Know

A knowledge-graph Q&A system built on a RAG knowledge base

Visit official site

What 语析Yuxi-Know is

The system is described as a knowledge-graph question-answering system built on a large-model RAG knowledge base, assembled with Llamaindex, VueJS, Flask and Neo4j, and able to handle knowledge-base Q&A, knowledge-graph retrieval and web search.

What you can do with it

  • Call models from OpenAI, major platforms in China or local vLLM
  • Import PDF, TXT and MD documents into the knowledge base
  • Run Neo4j-based knowledge-graph Q&A and show knowledge links
  • Add web search to bring webpage content into answers

Application scenarios

  • Staff query internal documents and a knowledge graph inside a company
  • Researchers manage literature and analyse relations with the graph
  • A customer-service solution answers questions with web search

Getting started

  • Clone the project code with Git to a local machine
  • Set the API_KEY for the service platform in an env configuration file
  • Start the project with Docker Compose for a development or production setup
  • Open the local page in a browser

Pros & cons

✓ What we like

  • Flexible model support and several knowledge-base formats
  • A single API_KEY for the chosen service platform is enough to start

! What to watch out for

  • Deployment requires Docker Compose and an env configuration file
  • Model calls depend on OpenAI, a Chinese platform or local vLLM

FAQ

What is it built with?

The material names Llamaindex, VueJS, Flask and Neo4j on a large-model RAG knowledge base.

Which models can it call?

OpenAI, major model platforms in China and local vLLM deployment are listed.

Last reviewed: 2026-11-15

More AI office assistant tools

View all →

How we review