Khoj
What is Khoj?
Khoj is an open-source personal AI assistant for searching, understanding, and creating content from your documents and the public web. It uses retrieval and language models to answer questions with relevant context from supported files, notes, and connected sources. Khoj is designed for individuals, researchers, developers, and knowledge workers who want a customizable second brain. Users can access it through web, desktop, Obsidian, Emacs, and WhatsApp interfaces. Its main differentiator is the option to self-host Khoj locally, supporting offline use and greater control over personal data.
How to use Khoj?
1. Install Khoj with Docker or Pip, or open an existing self-hosted instance; 2. Connect a supported language model and add documents, notes, or folders to index; 3. Ask questions, run research tasks, create an agent, or schedule an automation from the web or supported client.
Khoj's Core Features
Document Search: Find relevant passages across PDFs, Markdown, plaintext, Org-mode files, Notion pages, and other indexed sources.
Contextual Chat: Ask questions about your knowledge base or the public web and receive answers grounded in relevant context.
Web Research: Retrieve current publicly available information to support research-oriented conversations.
Custom Agents: Configure specialized assistants with distinct prompts, models, and task-oriented behaviors.
Scheduled Automations: Run recurring queries at selected times and receive the resulting research by email.
Code Execution: Generate and run simple Python code for data analysis, plots, and report creation.
Multi-Model Support: Connect online providers such as OpenAI, Anthropic, and Gemini or use compatible local models.
Self-Hosting: Deploy Khoj on personal hardware to retain control over indexed data and support private or offline workflows.
Multi-Client Access: Use Khoj through web, desktop, Obsidian, Emacs, or WhatsApp interfaces.
Image Generation: Create images from text prompts within supported Khoj conversations.
Khoj's Use Cases
- #1
Researchers searching across PDF papers, Markdown notes, and personal references with natural-language questions
- #2
Developers querying a private documentation library or GitHub repository while working in Obsidian or Emacs
- #3
Students using an AI tutor to understand course notes and retrieve relevant study material
- #4
Knowledge workers summarizing large document collections and drafting content from existing research
- #5
Privacy-conscious users building a locally hosted second brain without sending personal data outside their device
- #6
Individuals scheduling recurring research queries that produce email reports from current web information
Frequently Asked Questions
Analytics of Khoj
Monthly Visits Trend: Jun 2025 - Oct 2026
Traffic Sources
Top Regions
| Region | Traffic Share |
|---|---|
| United States | 46.95% |
| India | 10.90% |
| France | 9.05% |
| Russia | 7.46% |
| Germany | 6.08% |
Top Keywords
| Keyword | Traffic | CPC |
|---|---|---|
| khoj | 2.7K | $1.91 |
| khoj search engine | 130 | -- |
| стимулятор для пдд авто на русском языке пк как называется | 180 | -- |
| necesito algo como gems / proyectos que tengan buena memoria | 280 | -- |
| khoj-ai/khoj | 70 | -- |
Alternative of Khoj
ZeroTwo
A unified AI workspace for researching, creating, coding, and automating work across multiple models and connected apps.
LlamaIndex
LlamaIndex is an open-source framework for building AI-powered knowledge assistants by connecting large language models to enterprise data.
Dola
Dola is an all-in-one AI assistant for chatting, writing, research, summarization, learning, planning, voice interaction, and AI-powered image creation.
Tencent Yuanbao
Tencent Yuanbao is an advanced AI assistant powered by the Hunyuan large language model, offering smart chat, document analysis, image generation, and deep integration with the WeChat ecosystem.
Dust
Dust is a platform that empowers teams to build and deploy custom AI assistants using large language models to enhance productivity and automate workflows.
folk
A personal AI companion that remembers context, connects apps, monitors tasks, and helps with shopping through familiar messaging channels.
Make
Make is a no-code automation platform that enables users to connect apps, automate workflows, and streamline business processes without coding expertise.
Truecaller
Truecaller is a leading global caller ID and spam blocking platform that helps users identify unknown calls and messages while protecting them from fraud.