#2323rd of 42 in Agent platforms
yuxi
Multi-tenant agent platform with Milvus RAG, Neo4j graphs and LangGraph agents
Documentation ↗
(opens in a new tab)Website ↗
(opens in a new tab)Repository on GitHub ↗
(opens in a new tab)
- Stars
- 7.5k
- License
- MIT
- Last commit
- Oct 2026
- Last release
- Sep 2026
- Language
- Python
4,656 on 1 Apr 20267,451 on 21 Oct 2026
From GitHub's daily star counts.
Checked on GitHub.
Overview
Yuxi is a self-hosted, multi-user platform that combines document RAG (Milvus, MinerU/PaddleX/RapidOCR parsing, rerank), a Neo4j knowledge graph, and LangGraph-based agents with SubAgents, Skills, MCP and sandboxed file tools. Admins manage users, departments, model credentials and sharing scopes. The stack is Vue, FastAPI and PostgreSQL, started with Docker Compose.
Who it is for: Teams self-hosting a multi-user agent and knowledge platform with access control
Strengths
- Per-user, department and tenant permissions over knowledge bases, agents, Skills and models
- Built-in RAG evaluation and retrieval testing with embedding and rerank scores
- Supports MCP, Skills, SubAgents and sandboxed file output with in-browser preview
- Connects OpenAI, Anthropic, DeepSeek, Qwen and OpenAI-compatible endpoints such as vLLM
Weaknesses
- Heavy stack: Milvus, Neo4j, PostgreSQL, MinerU and PaddleOCR all run in Compose
- README is mainly Chinese; English version is a separate file
- Upgrading from v0.7.1 or v0.7.2 needs downtime, backup and migration
- Hardware requirements are not stated in the README
What it needs
- Docker + Compose
- Needs Docker Compose, Milvus, Neo4j, PostgreSQL, MinerU, PaddleOCR, LangGraph, LLM API
- Models: OpenAI, Anthropic, DeepSeek, Qwen, SiliconFlow
- port 5173
Also in Agent platforms
See all 42| Rank | Project | Score |
|---|---|---|
| 1 | LangflowVisual builder for AI agents and workflows, exposed as API or MCP server | 81 out of 100 |
| 2 | daily_stock_analysisDaily LLM stock analysis for A-share, HK, US and other markets, pushed to chat apps | 81 out of 100 |
| 3 | resume-matcherTailors resumes and cover letters to job descriptions using local or hosted LLMs | 81 out of 100 |
| #4 | vibe-tradingFinance research agent for backtests, market data and broker connectors | 78 out of 100 |
| #5 | deer-flowAgent harness that runs sub-agents, memory and sandboxed skills for long tasks | 77 out of 100 |