MaxKB vs rag-agent-platform

Self-hosted RAG and agent platforms compared

MaxKB and rag-agent-platform both provide self-hosted applications for building knowledge-based LLM agents with document retrieval and MCP support. MaxKB emphasizes a broader agent-building toolkit and integrations, while rag-agent-platform combines retrieval and agents with a Java and Next.js stack and more operational infrastructure to manage.

MaxKBrag-agent-platform
LanguagePythonJava
LicenseGPL-3.0MIT
Stars22.9k89
Forks3.2k7
Last analyzedOct 3, 2026Oct 14, 2026

Key differences

  • MaxKB includes workflows, a function library, and integration with third-party business systems; rag-agent-platform focuses on document retrieval, configurable agents, and MCP gateway connections.
  • MaxKB is written in Python and uses Vue.js, Django, LangChain, PostgreSQL, and pgvector; rag-agent-platform uses Java with a Next.js frontend and PostgreSQL with PGVector.
  • MaxKB supports text, image, audio, and video input and output; rag-agent-platform documents streamed conversations using Server-Sent Events and asynchronous document processing through RabbitMQ.
  • MaxKB is licensed under GPL-3.0, while rag-agent-platform uses the MIT license.
  • rag-agent-platform requires services including RabbitMQ, S3-compatible storage, model and reranker providers, and SMTP; MaxKB's documented stack centers on PostgreSQL and pgvector alongside its application components.
  • MaxKB has 22.9k stars and 3.2k forks; rag-agent-platform has 89 stars and 7 forks. Its supplied project notes also describe development concentrated in one contributor and no releases recorded in the past year.

Choose MaxKB if you…

  • need workflows, a function library, and tools for more involved agent tasks.
  • want to connect knowledge-based Q&A to existing business systems or use several public and private model providers.
  • need support for text, image, audio, and video input and output.
Read the MaxKB analysis →

Choose rag-agent-platform if you…

  • want a Java backend and Next.js frontend for a self-hosted RAG and agent application.
  • can operate PostgreSQL, RabbitMQ, and S3-compatible storage alongside the required model and other external services.
  • are prepared to configure MCP gateways and tool containers manually.
Read the rag-agent-platform analysis →

Compare the numbers side by side, or add more projects

This comparison is generated with AI from the OSRepos analyses of both projects. Always check each project's repository and documentation before choosing.