What is the project about?
MaxKB (Max Knowledge Base) is an open-source chatbot system built on Large Language Models (LLMs) and Retrieval-Augmented Generation (RAG). It's designed to create a knowledge-based question-answering system. It is presented in both English and Simplified Chinese.
What problem does it solve?
It addresses the challenge of creating intelligent Q&A systems that can leverage existing knowledge sources (documents, websites, etc.). It specifically aims to reduce the "hallucinations" (inaccurate or nonsensical answers) that LLMs can sometimes produce by grounding the responses in retrieved information. It also simplifies integrating AI-powered Q&A into existing systems.
What are the features of the project?
- Ready-to-Use: Supports direct document uploads and automatic web crawling. Handles text splitting, vectorization, and RAG automatically.
- Flexible Orchestration: Includes a workflow engine and function library for creating custom AI processes.
- Seamless Integration: Designed for easy integration with third-party systems (no coding required).
- Model-Agnostic: Works with various LLMs, both private (DeepSeek, Llama, Qwen) and public (OpenAI, Claude, Gemini).
What are the technologies used in the project?
- Frontend: Vue.js
- Backend: Python/Django
- LLM Framework: LangChain
- Database: PostgreSQL with pgvector (for vector storage and similarity search)
- Containerization: Docker
What are the benefits of the project?
- Improved Accuracy: RAG reduces LLM hallucinations, leading to more reliable answers.
- Flexibility: The workflow engine allows customization for various business needs.
- Easy Integration: Quickly adds intelligent Q&A to existing systems.
- Model Choice: Users can select the LLM that best suits their requirements.
- Simplified Knowledge Management: Streamlines the process of creating and using a knowledge base.
What are the use cases of the project?
- Intelligent Customer Service: Providing automated answers to customer inquiries.
- Corporate Internal Knowledge Bases: Making it easier for employees to find information.
- Academic Research: Supporting research by providing quick access to relevant information.
- Education: Creating interactive learning tools.
- Any scenario requiring question answering based on a specific set of documents or knowledge.
