Own your intelligence.
A private AI assistant that runs entirely on your computer. No accounts, no API keys, no token limits.
On-device AI that works
No complex setup.
Preconfigured for productivity.
Every document becomes part of your agent's knowledge base with no data leaving your device.
Learn more about Document KnowledgeMeeting Assistant
Every call becomes context.
Automatically transcribes and summarizes any meeting entirely on your computer. Action items, key decisions, and full transcripts ready the moment you hang up. No bots joining your meeting. No cloud processing.
- No bot joins your call — listens locally
- Auto-generated summaries & action items
- Works with any meeting platform
Magic Features
Intelligence at your mouse pointer.
Speak and your words not only appear, but are refined in real time - and it knows what you're looking at for precise answers. Fully on device.
Learn more about Magic EchoUp and running in a few clicks.
Download the app
One file. No dependencies. Works on macOS, Windows, and Linux.
Pick a model (or let us pick)
Select your preferred model or let AnythingLLM auto-recommend the best one for your hardware specs and our experience.
You're ready to go
Chat with docs, do research, build workflows, and more — all running locally and on device. No cloud API key needed.
Scale to infinity
The custom tool builder and developer API make it easy to fit AnythingLLM to your exact needs - no extra bloat.
Also on Android
The same agent, in your pocket.
AnythingLLM Mobile runs small models fully on-device, with web search, document chat, file generation, scheduled jobs and memory built in. Pair it with your desktop to reach larger models and your full workspaces from anywhere.
See AnythingLLM MobileBlog
AnythingLLM and You.com bring real-time web search to everyone
You.com's accurate, fresh, and fast web search is now the default in AnythingLLM. Web search is on from your first chat, with no API key and nothing to setup.
AnythingLLM now runs Microsoft Foundry Local on every Windows PC
Starting in v1.16.1, AnythingLLM can tap into Foundry Local to run models on the CPU, GPU, and NPU of any Windows machine with no extra installation.
Your laptop can take meeting notes now. Stop paying someone else to do it.
AnythingLLM's Meeting Assistant records, transcribes, and summarizes every meeting on your own computer, indexes it into your context, and never charges you. Here is why we built it that way.
MIT Licensed. Open Source.
Built in the open.
Trusted by millions.
AnythingLLM is fully open-source and licensed for organizational use. Self-host a multi-user deployment on your own servers, or just use the desktop app. Either way, you don't rent your intelligence and your data is yours.
Don't rent intelligence. Own it.
Break free from rate-limits and token costs. Download AnythingLLM and get a privacy-focused, extensible, and capable AI agent running on your device in minutes.
Download AnythingLLM — Free