MaxKB is an open-source AI assistant designed for enterprise use, offering a ready-to-use RAG (Retrieval-Augmented Generation) intelligent Q&A application. It targets scenarios like intelligent customer service, internal knowledge bases, and academic research, providing a seamless way to integrate AI into existing business systems.

How It Works

MaxKB leverages LangChain for its RAG pipeline, enabling document uploading and web crawling with automatic text splitting and vectorization. This approach aims to reduce large language model hallucinations and enhance Q&A accuracy. It also features an agentic workflow engine with MCP tool-use capabilities for orchestrating complex AI processes.

Quick Start & Requirements

• Install: docker run -d --name=maxkb --restart=always -p 8080:8080 -v ~/.maxkb:/var/lib/postgresql/data -v ~/.python-packages:/opt/maxkb/app/sandbox/python-packages 1panel/maxkb
• Access: http://your_server_ip:8080 (admin/MaxKB@123)
• Prerequisites: Docker. Offline installation documentation is available for users experiencing pull issues.
• Resources: Requires Docker and sufficient system resources for container operation.

Highlighted Details

• Supports multi-modal input/output (text, image, audio, video).
• Model-agnostic, supporting both private (DeepSeek, Llama, Qwen) and public (OpenAI, Claude, Gemini) LLMs.
• Built upon LangChain, Django, and PostgreSQL with pgvector.
• Offers zero-coding integration into third-party business systems.

Maintenance & Community

The project is actively developed by 1Panel-dev. Further community or roadmap links are not explicitly provided in the README.

Licensing & Compatibility

Licensed under the GNU General Public License v3.0 (GPLv3). This is a strong copyleft license, meaning derivative works must also be open-sourced under GPLv3. Commercial use or integration into closed-source systems may require careful consideration of license obligations.

Limitations & Caveats

The GPLv3 license imposes significant restrictions on proprietary use. While it supports RAG and agentic workflows, it is positioned as a ready-to-use application rather than a middleware platform for building custom LLM applications.