yuezhiai/jonex

All-in-One Multimodal Parsing Engine + Ontology-Powered, LLM Wiki-Driven AI-Ready Knowledge Engine

What it solves

Jonex addresses the challenge of turning raw, multimodal enterprise data—such as documents and videos—into structured, actionable knowledge. It bridges the gap between raw information and AI-ready intelligence by organizing data into a governed system that supports reasoning, traceability, and retrieval.

How it works

Jonex functions as an end-to-end knowledge platform that combines a multimodal parsing engine with a knowledge engine. It ingests raw content, parses it, and compiles it into a domain-specific ontology (a structured map of concepts and relationships). This data is stored using both vector and graph indexing. When a user asks a question, the system performs source-grounded retrieval, allowing AI agents to access structured knowledge with verified references and reasoning traces.

Who it’s for

Enterprises that need to build AI agents capable of performing complex, domain-specific tasks. It is designed for organizations that require a governed, multi-tenant platform to manage large volumes of internal data and transform them into reusable knowledge services.

Highlights

  • Multimodal Parsing: Automatically processes documents, audio, and video into structured data.
  • Ontology-First Retrieval: Uses graph-based knowledge structuring to improve the accuracy and reasoning capabilities of AI agents.
  • Source Traceability: Provides clear references for all retrieved information, ensuring answers are grounded in actual data.
  • Enterprise-Ready: Built with multi-tenant support and a unified API gateway for secure, scalable integration.
  • Flexible Infrastructure: Supports various integrations, including Neo4j for graph data, Milvus for vector storage, and OpenAI-compatible LLM endpoints.

Related

  • Project
    java-up-up/nexus-agent

    An enterprise-grade AI agent platform that implements a complete loop from document governance and hybrid RAG retrieval to ReAct-based agentic reasoning and MCP tool integration.

  • Project
    onyx-dot-app/onyx-foss

    Onyx is a self-hosted knowledge and context layer that connects to over 50 applications to provide AI agents with secure, relevant access to your organization's internal data.

  • Project
    opendatalab/MinerU

    A high-fidelity document parsing tool that converts PDFs, images, and Office files into structured Markdown or LaTeX, featuring multi-tiered quality levels and stable locators for AI agents.

  • Project
    zj-unicom-ai/UniEmployee

    UniEmployee is an enterprise-grade platform for building and managing AI digital employees by codifying business processes, professional expertise, and safety guardrails into observable, multi-step workflows.

  • Project
    aigc-apps/PAI-RAG

    An open-source Agentic RAG framework that combines multimodal understanding, ReAct-based agents, and enterprise-grade knowledge management for complex intelligent Q&A.