Y

Yuxi

A privately deployable, multi-tenant knowledge intelligence platform supporting RAG, knowledge graphs, and multi-agent orchestration.

🇨🇳 DomesticFreeOpen sourceGeneral开源知识图谱RAG多租户MCP/Skills沙盒
Platforms: WebAPI
Region
Domestic
Pricing
Free
Open source
Yes
GitHub Stars
★ 6.5k
Source
GitHub
Added
2026-08-19
Last verified
2026-08-19
Yuxi

Overview

Yuxi (YuXi) is a privately deployable, multi-tenant knowledge intelligence platform integrating RAG retrieval, Milvus-powered knowledge graph, LangGraph-based multi-agent orchestration, MCP/Skills, sandbox tools, and permission management. Administrators can onboard models, build knowledge bases, and configure user and department-level permissions. Users can invoke knowledge, tools, and sub-agents through a unified chat interface, receiving answers with source citations, context-aware reasoning based on the knowledge graph, and previewable, downloadable outputs. Yuxi is ideal for team collaboration, offering multi-tenancy, user/department permissions, centralized model configuration, external API key integration, and Docker Compose deployment out of the box.

Key features

  • Collaboration between knowledge and agents
  • Support for long-running tasks and deliverable outputs
  • Multi-tenancy and permission management
  • Docker Compose ready-to-use deployment

Use cases

Enterprise internal knowledge managementTeam collaboration and task executionIntelligent customer service systemsAutomated document processing

Pros

  • Strong knowledge graph support
  • Flexible multi-agent orchestration
  • Comprehensive permission management system
  • Easy deployment and scalability

Limitations / notes

  • Requires some technical expertise
  • Slow cold start performance

Who it's for

DevelopersEnterprise teamsResearchersIT administrators

This overview was compiled by AI from public sources and may contain inaccuracies — please refer to the official site.

FAQ

Is it free?

Yes, this project is fully open-source and free under the MIT license.

Do I need to use scientific internet access?

No, but you will need at least one large model API compatible with the OpenAI interface.

Does it support Chinese?

Yes, the project is primarily in Chinese, and the development team is based in China.

Can it be used commercially?

Yes, it is licensed under MIT, allowing commercial use.

Similar agents

Something wrong? Let us know on the About page and we'll fix it.