anything-llm — AnythingLLM is an all-in-one AI application for building a private, local-first ChatGPT experience with document ingestion and AI agent capabilities.

AnythingLLM is an all-in-one AI application for building a private, local-first ChatGPT experience with document ingestion and AI agent capabilities.

About this project

AnythingLLM is a comprehensive AI application that allows users to create a private ChatGPT-like environment. It supports connecting to various local or cloud LLMs, ingesting documents (PDF, TXT, DOCX, etc.), and utilizing AI agents for complex workflows. Key capabilities include: - Dynamic Model Routing and intelligent skill selection for AI agents. - Support for a wide array of LLM providers (OpenAI, Anthropic, Google Gemini, Ollama, LM Studio, etc.). - Integration with multiple vector databases such as LanceDB, Pinecone, Milvus, and Weaviate. - Multi-user support and permissioning (available in the Docker version). - Built-in tools for audio transcription (STT) and text-to-speech (TTS). - A no-code AI Agent builder and MCP compatibility. - A custom embeddable chat widget for websites. The project is structured as a monorepo containing a React frontend, a NodeJS express server, and a document collector.