AnythingLLM is a document-and-agent app built around RAG: create workspaces, drop in files and chat against them with an embedded vector store, plus an agent mode.
| Category | Self-hosted chat UI |
| Type | Desktop / Docker app |
| License | MIT |
| Runs locally | Self-hosted |
| Built with | JavaScript |
| Skill level | Beginner |
| Best for | teams whose main job is chatting with documents |
Other open-source self-hosted chat ui tools worth comparing:
Open WebUIChatGPT-style interface over any model
LibreChatMulti-provider open-source ChatGPT clone
Lobe ChatModern, design-forward chat framework
big-AGIPro-grade chat for many models
ChatboxDesktop client for many AI models
NextChatLightweight ChatGPT-style UI, deploys in one click
Chatbot UIClean open ChatGPT clone with files and profiles
HuggingChat UIThe interface behind HuggingChat, self-hostable
KhojYour AI second brain over your notes
KotaemonClean open UI to chat with documents
Open WebUI PipelinesAdd custom logic to any OpenAI-compatible UIAnythingLLM is free and open-source (MIT license), so you can use, self-host and modify it at no cost.
Yes. AnythingLLM is designed to run on your own machine or server, keeping your data private.
Popular open-source alternatives include Open WebUI, LibreChat, Lobe Chat. See the comparisons above to choose.
Browse the full directory of open-source AI tools, models and projects — updated daily.
Browse all tools →