
In simple English
This tool acts as a personal research assistant that connects to your own files and notes. You can ask questions and get answers drawn straight from your private documents. It lets you chat with different artificial intelligence models, both online and offline. You can run it on your own server for total privacy or use the hosted web version. It is free and open-source, though running it locally needs a powerful computer.
What it is
Khoj is a self-hosted artificial intelligence platform that indexes your notes, PDFs, and documents to answer questions through semantic search (a way for a program to understand the meaning of your words rather than just matching exact keywords). It connects to popular text models, web search, and chat apps like WhatsApp.
What it is good for
Solo builders and researchers who want an intelligent search engine over their personal knowledge base without sending private notes to third-party servers.
- Searches across your local files, notes, and the internet at the same time.
- Connects to many different language models, including local options that keep your data private.
- Accessible through a web browser, desktop apps, and chat tools like WhatsApp.
- Requires Docker (a way to run a program inside its own sealed box, so it cannot break anything else on your server) or a Python setup to run yourself.
- Running local language models requires a computer with a powerful graphics card and plenty of memory.
- Uses the AGPL-3.0 licence, which has strict rules if you modify and host it as a commercial web service.
Manual file searching and commercial hosted chat assistants that require uploading private notes to the cloud.
Obsidian for note-taking and Docker for running services locally.
- LibreChat-AI/LibreChat46k stars
Provides a multi-model chat interface with support for local and online artificial intelligence providers.
- open-webui/open-webui156k stars
Offers a feature-rich self-hosted web interface for interacting with local language models and documents.
- ggml-org/llama.cpp132k stars
A C and C++ library and command-line tool for running language models locally.
- EleutherAI/lm-evaluation-harness14k stars
Runs standard question sets against a language model and reports the scores.
- mudler/LocalAI50k stars
Serves text, vision, speech, image and video models from one self-hosted service that imitates the paid interfaces.
docker pull ghcr.io/khoj-ai/khoj docker run -p 8000:8000 ghcr.io/khoj-ai/khoj