AnythingLLM
Free tierThe all-in-one AI application — chat with docs, run AI agents, fully private and local
Key strengths
Facts verified against anythingllm.com on
Recent changes
- Models updated: Bring your own LLM API key; built-in support for llama.cpp (via AMD Lemonade), Microsoft Foundry Local, and others
- Starting price: $0/mo → $50/mo
- Plans updated: Desktop app is free. Cloud/hosted plans start at $50/mo (Basic), $99/mo (Pro), and Enterprise (contact for quote). Self-hosting with Docker is free.
Get alerts when AnythingLLM changes
Free. We'll email you when pricing, the free tier, API access or open-source status changes.
AnythingLLM is an MIT-licensed, open-source AI assistant platform designed for fully local, on-device inference with zero data egress. It supports a bring-your-own-LLM model, with built-in integrations for llama.cpp (via AMD Lemonade), Microsoft Foundry Local, and other local or API-backed providers. The desktop app ships as a free, cross-platform binary (macOS, Windows, Linux, Android), while self-hosted multi-user deployments are supported via Docker at no cost. Cloud/hosted plans are available starting at $50/mo (Basic), $99/mo (Pro), and Enterprise (contact for quote).
Key architectural capabilities include:
- RAG (Retrieval-Augmented Generation) via a built-in document knowledge base
- Custom agent skills for extensible, task-specific automation
- Developer API for programmatic integration
- Local meeting transcription and summarization with no third-party calls
- Self-hostable multi-user deployment for team or enterprise rollouts
- 67,000+ GitHub stars, reflecting strong community adoption since its 2023 founding