AI Chatbots
Self-hosted chatbots and agent builders trained on your own content.
AI Chatbots
Self-hosted chatbots keep conversations and knowledge on your infrastructure. The projects here range from embeddable website assistants to visual builders for multi-channel agents.
Compare them by how they ingest content, which channels they support, and whether they include retrieval, analytics, and human handoff out of the box.
6 projects
- langgenius/dify — Visual platform for building LLM apps with RAG pipelines, agents and model integrations — Multi-channel support: No; Visual flow builder: Yes; LLM integration: Yes; Human handoff: No; Knowledge base answers: Yes; MCP tool integration: Yes; Embeddable chat widget: Yes
- open-webui/open-webui — Self-hosted web UI for chatting with Ollama and OpenAI-compatible models offline — Multi-channel support: No; Visual flow builder: No; LLM integration: Yes; Human handoff: No; Knowledge base answers: Yes; MCP tool integration: Yes; Embeddable chat widget: No
- hexabot-ai/Hexabot — AI workflow automation platform combining agents, actions and chat channels — Multi-channel support: Yes; LLM integration: Yes
- hexastack/hexabot — Hexabot is an open-source AI chatbot / agent builder. — Multi-channel support: Yes; Visual flow builder: Yes; LLM integration: No; Human handoff: No; Knowledge base answers: Yes; MCP tool integration: Yes; Embeddable chat widget: Yes
- botman/botman — Framework-agnostic PHP library for building bots across chat platforms — Multi-channel support: Yes; Visual flow builder: No; Knowledge base answers: No; MCP tool integration: No; Embeddable chat widget: No
- raksbisht/sitechat — Open-source, self-hosted AI chatbot trained on your website and docs. RAG, embeddable widget, no vendor lock-in. — Multi-channel support: No; Visual flow builder: Yes; LLM integration: Yes; Human handoff: Yes; Knowledge base answers: Yes; MCP tool integration: No; Embeddable chat widget: Yes