AnythingLLM is an open-source AI application for chatting with documents, building private knowledge workspaces, and running AI assistants with a wide choice of language models. It combines document ingestion, retrieval-augmented generation, conversational interfaces, agent capabilities, and model-provider flexibility in a desktop or self-hosted environment. Users can organize files and knowledge into separate workspaces, connect local or cloud models, and ask questions grounded in their own content.

Beyond document chat, AnythingLLM supports agent tools, multi-user deployments, vector databases, embedding providers, and integrations intended for both individual and team use. Its local-first options make it particularly attractive for users who want more control over sensitive documents or want to run models on their own infrastructure. Overall, AnythingLLM provides a relatively complete interface for turning private files and external AI models into a customizable internal assistant without building a RAG application from scratch.