What it does
Core capabilities at a glance
- Search across markdown, PDF, images, and code files
- Natural language Q&A over your personal data
- Automated knowledge graph construction
- Plugs into Obsidian, Emacs, and desktop
- Self-hostable with Docker
- Works with Ollama, OpenAI, and any OpenAI-compatible API
Deep dive
The full breakdown - performance, comparisons, and setup
Khoj
Khoj is your personal AI search engine that works on your own data. It indexes your markdown notes, PDFs, images, and code files, then lets you ask questions in natural language.
What it is
Khoj is a self-hostable AI search application developed by Debanjum Singh. It focuses on personal knowledge management - connecting to Obsidian, Emacs, and your file system - and building a knowledge graph of your information automatically.
How it stacks up
| Khoj | Onyx | AnythingLLM | RAGFlow | |
|---|---|---|---|---|
| Personal focus | ✓✓ | Enterprise | General | Document |
| Obsidian integration | ✓ | ✗ | ✗ | ✗ |
| Knowledge graph | ✓ | ✗ | ✗ | ✓ |
| Desktop app | ✓ | ✗ | ✓ | ✗ |
| Best for | Personal PKM | Enterprise search | General RAG | Complex docs |
Get started
docker run -d -p 42110:42110 \
-e OPENAI_API_KEY=sk-your-key \
ghcr.io/khoj-ai/khoj:latestWhen to use something else
- Enterprise with connectors: Onyx for Slack/Confluence/Drive
- Better document parsing: RAGFlow for complex PDFs
- Desktop document chat: AnythingLLM or GPT4All for local-only
From the community
What users are saying
“Likewise! That was one of the impulses behind working on Khoj -- we have all this data about places we go to, things we do, websites we traffic, but such poor tooling into how to actually retrieve that information right now.”
“Hello! One of the developers of Khoj here. The way we see it, building in the open is going to be critical for creating an aligned, trustworthy AI assistant. Note: while all LLM tools look fairly similar on the surface these days, our specific approaches are fairly different. Give us a try and see what you think :-)”
“Might look into some of the tools like novoids Memacs. Notion here is to build tools that push feeds, history data, into Emacs. Using org in your use case with the Khoj tool, could be the "glue" you need to tie it all together. https://github.com/novoid/Memacs#readme .”
Frequently asked
Quick answers to common questions
What is Khoj?
Khoj is a rag tool for local AI workloads. Open-source, personal AI assistant that searches your notes, documents, and knowledge graph.
Is Khoj free and open source?
Yes, Khoj has 37,738 GitHub stars and is licensed under AGPL-3.0. You can self-host it for free on web, docker, macos, linux, windows.
What platforms does Khoj support?
Khoj runs on web, docker, macos, linux, windows.
What hardware do I need for Khoj?
The hardware requirements depend on which models you run. Check our hardware directory for compatible GPUs and systems. Khoj has 37,738 GitHub stars and an active community.
Does Khoj support GPU acceleration?
Khoj supports GPU acceleration via CUDA, Metal, or Vulkan depending on your platform. For the best performance, pair it with an NVIDIA RTX 4090 or 5090.
What are the best alternatives to Khoj?
Popular alternatives include other rag tools in our directory. Browse our full collection at /tool for comparisons, community reviews, and benchmark data to find the right fit for your workflow.
How much does Khoj cost?
Khoj is free-tier. There is a free tier available with paid upgrades for advanced features.
Pairs well with
Complementary tools, models, and hardware