Open-source AI Tools
AnythingLLM

An all-in-one desktop and enterprise application for building private RAG systems with any LLM.

Use tool
Use Case
Creating isolated, private AI workspaces for different company departments using local files and custom LLM providers.
Website Preview
AnythingLLM website preview

AnythingLLM is a comprehensive, all-in-one application engineered to turn any collection of documents, links, or media into a fully private, conversational knowledge base. Available as both a lightweight desktop client and a multi-tenant enterprise server, it provides complete flexibility over your local AI setup by decoupling models, embedding engines, and vector databases.

The application features advanced workspace isolation, allowing users to organize different sets of documents for specific projects. AnythingLLM includes built-in web scraping, automated document parsing, robust user management, and seamless integration with both commercial APIs and local open-source models like Ollama or LM Studio, creating a completely secure environment for enterprise knowledge management.

Relevant Sites

Jan

An open-source, privacy-first desktop alternative to ChatGPT that runs fully offline on your computer.
Zilliz

A fully managed, enterprise-grade cloud service built on top of the powerful Milvus vector database.
Hugging Face

The community-driven platform for open-source machine learning models.
Milvus

An open-source vector database built to manage billion-scale vector datasets for AI and similarity search.
llama.cpp

A high-performance C++ implementation for running LLMs on consumer hardware.
LocalAI

An open-source, drop-in replacement REST API for OpenAI, running models locally on your own infrastructure.