HubLensLLMkhoj-ai/khoj
khoj-ai

khoj

AI#LLM#Semantic Search#Self-hosted#Personal Assistant
View on GitHub
34,330

// summary

Khoj is a versatile personal AI application designed to extend your capabilities by integrating with various local and online large language models. It allows users to interact with their personal documents and the internet through a unified interface accessible across multiple platforms. The project is open-source and supports flexible deployment options ranging from private on-device setups to scalable enterprise cloud solutions.

// technical analysis

Khoj is a versatile personal AI application designed to function as an extensible 'second brain' that scales seamlessly from local, on-device usage to cloud-based enterprise deployments. By integrating various LLMs with personal document repositories, it solves the problem of fragmented information management and repetitive research tasks. The project prioritizes flexibility and privacy, offering a modular architecture that supports custom agents and multiple access points, including web, desktop, and mobile platforms.

// key highlights

01
Supports a wide range of local and online LLMs, including Llama3, GPT, Claude, and DeepSeek, for flexible conversational AI.
02
Enables semantic search and retrieval across diverse file formats like PDF, Markdown, Notion, and Word documents.
03
Allows the creation of custom AI agents with specific knowledge bases, personas, and tools to handle specialized tasks.
04
Provides multi-platform accessibility, allowing users to interact with their AI via browser, Obsidian, Emacs, desktop, mobile, or WhatsApp.
05
Automates research workflows by delivering personalized newsletters and smart notifications directly to the user's inbox.
06
Offers full self-hosting capabilities for users who prioritize data privacy and local control over their AI infrastructure.

// use cases

01
Chat with local or online LLMs using custom knowledge bases
02
Perform advanced semantic search across PDFs, Markdown, Notion, and other document formats
03
Automate research tasks and receive personalized notifications

// getting started

To begin using Khoj, you can either access the hosted cloud version immediately at app.khoj.dev or follow the self-hosting instructions provided in the official documentation. Developers should visit the documentation site to review the setup guide and explore the contributing guidelines if they wish to participate in the project's development.