Home
Softono

Lingxi Core

Open source Apache-2.0 Python
12
Stars
4
Forks
5
Issues
2
Watchers
8 months
Last Commit

 About Lingxi Core

无需编程,即可快速构建基于先进AI模型的各类智能应用——从简单的问答机器人到包含复杂逻辑的工作流。构建完成后,一键发布 至主流社交平台、Web网页,或通过开放API无缝集成到现有系统进行深度开发。

Platforms

Web Self-hosted

Languages

Python

Need Help Installing Lingxi Core?

We provide expert installation service for this software. Our team will install, configure, and secure Lingxi Core on your server. plans start at just $30.

Lingxi Core

View on GitHub

Read this in other languages: English, 中文.

《AI Application Development Platform》

Build Intelligent Apps with Zero Code, One-Click Publishing, Infinite Scalability

Platform Overview

No coding required—rapidly build various intelligent applications powered by advanced AI models, from simple Q&A bots to complex logic-driven workflows. Upon completion, publish with one click to major social platforms, web pages, or integrate seamlessly into existing systems via Open APIs for deeper development.

Technical Foundation

  • Frontend: Vue3 (Reactive Framework) + Arco Design ( Enterprise-Grade UI) + Tailwind CSS (Efficient Styling)
  • Backend: Python 3.12 + Flask (Lightweight Web Framework) + LangChain 0.3 & LangGraph 0.2 (Core AI Orchestration) + Redis (Cache/Message Queue) + JWT (Secure Authentication)
  • Data Storage: PostgreSQL (Relational Database) + Weaviate (High-Performance Vector Database)
  • Asynchronous Processing: Celery (Distributed Task Queue)

Core Feature Highlights

  1. Intelligent Homepage: Built-in assistant Agent helps you quickly start creating personalized AI applications.
  2. Personal Workspace: Centrally manage your AI applications, plugins, workflows, and knowledge base resources.
  3. Powerful AI Application Orchestration:
    • Custom AI Personas & Logic: Flexibly define AI personalities, response strategies, and optimize prompts.
    • Multi-Agent System (Beta): Supports both single-Agent operation and Supervisor-coordinated multi-Agent systems.
    • Rich Capability Extensions: Integrate plugins, workflows, knowledge bases; configure long-term memory, opening prompts/preset questions, user question suggestions, multi-language/multi-modal input/output, content moderation, etc.
    • MCP Client Configuration: Adhering to the MCP protocol specification, it can seamlessly invoke external tools and services, enhancing external scalability.
    • Real-Time Preview & Debugging: Test while you build with WYSIWYG editing, boosting development efficiency.
    • Model Flexibility: Supports custom LLM integration and key parameter tuning.
    • Version Control: Enables historical version rollback for AI applications, ensuring configuration security.
  4. Effortless Application Publishing: Offers one-click publishing to Web interfaces and WeChat Official Accounts.
  5. Multi-Dimensional Analytics: Provides comprehensive operational insights for Agent applications (e.g., interaction volume, effectiveness).
  6. Custom Plugin Development: Extend platform capabilities flexibly to meet custom business requirements.
  7. Visual Workflow Orchestration:
    • Rich node library: Start/End, LLM Invocation, Plugin Execution, Knowledge Retrieval, Template Transformation, Intent Recognition, Loop Iteration, HTTP Request, Python Code Execution, Conditional Branching, etc.
    • Supports workflow debugging to ensure logical accuracy.
  8. Intelligent Knowledge Base Management:
    • Supports multi-format document upload, intelligent text segmentation, and preprocessing.
    • Leverages Weaviate vector database for RAG (Retrieval-Augmented Generation) Enhanced Retrieval: Supports multiple efficient retrieval strategies like similarity search, full-text search, hybrid search.
  9. Built-In Resource Hub: Offers ready-to-use pre-built application templates and curated plugin libraries to accelerate application development.
  10. Open API Interface: Provides comprehensive Open APIs to directly invoke Agent applications, supporting both streaming and non-streaming outputs for deep integration.