AI · Early access

AnythingLLM hosting on your own server

A local-first chat app for talking to your own documents.

Early access. Join the list and we email you when AnythingLLM opens on VPS.

What AnythingLLM does

AnythingLLM is a retrieval-augmented chat application that lets users upload documents and query them through a local or remote LLM, keeping data on infrastructure they control. It supports multiple vector databases and LLM providers, and is aimed at small teams wanting private document search. It also supports multi-user workspaces, so a small team can share the same document collections and chat history.

AnythingLLM at a glance

License: MIT
Source code: github.com/Mintplex-Labs/anything-llm
Website: anythingllm.com
Runs on: Your own server (VPS)
Good to know: Needs Node.js and a vector store; a GPU helps only if it is also running a local embedding or LLM model.

Private document search and Q&A
Internal knowledge base chat
Local RAG experimentation

AnythingLLM, connected to the rest of your business

On your domain
AnythingLLM answers at an address like anythingllm.yourbusiness.com, with SSL and daily backups switched on.
Mail from your address
AnythingLLM sends its emails from your CloudWish business mailbox. Business email
AI with a cap you set
Give AnythingLLM its own AI key with a monthly spend cap. Usage shows as its own line on your bill. AI in your apps (early access)