Chat with your own documents and run local LLMs with zero setup RAG, AI agents, and privacy-first design bundled into one desktop app.
Download AnythingLLMOpens the official anythingllm.com download page ยท Windows, macOS & Linux
Maker
Mintplex Labs
Cost
Free
Runs
Fully offline
Local engine
Built-in Ollama
Drop in PDFs, spreadsheets, or notes and ask questions grounded in that content (RAG).
Ships with Ollama under the hood โ a model downloads and runs with no separate setup.
Swap in GPT, Claude, or Mistral via API key if you'd rather use a cloud model for some chats.
Not a SaaS product โ models, documents, and chat history stay on your machine by default.
Click the download button above
It opens anythingllm.com's real desktop download page โ pick the Windows x86-64 or ARM64 build.
Allow the browser/SmartScreen prompt
Click "Keep" if your browser flags it, then "More info" โ "Run anyway" if Windows SmartScreen appears โ normal for less common installers, not a sign of a problem.
Install for "Current User" only
Don't choose "install for all users" โ the developers note this causes issues and isn't supported.
Launch and pick a model
First boot takes a minute or two while it sets up the local model โ later launches are near-instant.