Mintplex-Labs/anything-llm

What is it?

Key features

• 01Chat with uploaded PDF, TXT, DOCX and other documents, with source citations
• 02Built-in local LLM engine supporting CPU and GPU, plus connections to many cloud and local model providers
• 03No-code AI agent builder and custom agents with tool use and web browsing
• 04Scheduled tasks that run prompts or agent workflows on a cron schedule
• 05Multi-user support with per-user permissions in the Docker version
• 06Developer API and embeddable chat widget for custom integrations

Use cases

• Building a private question-answering assistant over internal company documents
• Running a local-first AI chat setup on your own machine without sending data to cloud services
• Giving a team shared AI workspaces with separate access controls
• Embedding a chat assistant on a website or integrating AI chat into an existing product via API

Quick start

1. 1

   Clone the repository

2. 2

   Fill in the required .env files from the repo root

   yarn setup
3. 3

   Start the server (from repo root)

   yarn dev:server
4. 4

   Start the frontend (from repo root)

   yarn dev:frontend

Things to know

• !Development setup requires filling in several .env files and needs Yarn.
• !Multi-user support and the embeddable chat widget are available only in the Docker version.
• !Telemetry collects anonymous usage information by default.

Topics

FAQ

What is anything-llm?

All-in-one AI app for chatting with your documents and running AI agents, with local-first setup, multi-user support and many LLM providers. AnythingLLM is a self-hostable AI application for building a private ChatGPT-style assistant over your own documents. You connect a local model (through its built-in llama.cpp-based engine, Ollama, LM Studio and others) or a cloud LLM provider such as OpenAI, Anthropic or Gemini, then upload files to create workspaces that the model can answer questions from.

Is anything-llm free to use? What license does it have?

anything-llm is open source under the MIT license. Check the license terms before using it in commercial products.

Is anything-llm still maintained?

Actively maintained. The last commit to anything-llm was on Oct 17, 2026.

What can I use anything-llm for?

Building a private question-answering assistant over internal company documents Running a local-first AI chat setup on your own machine without sending data to cloud services Giving a team shared AI workspaces with separate access controls Embedding a chat assistant on a website or integrating AI chat into an existing product via API

How do I install anything-llm?

Clone the repository Fill in the required .env files from the repo root: yarn setup Start the server (from repo root): yarn dev:server Start the frontend (from repo root): yarn dev:frontend

Where can I get anything-llm?

The source code and installation instructions are on GitHub: https://github.com/Mintplex-Labs/anything-llm