Personalized AI assistant for university students who answers from the students' slides, textbooks, and notes.
MIT License
Dahih Al-Dofaa ๐ is an open-source, desktop application designed to revolutionize the way students approach learning. It's a groundbreaking, offline study companion ๐งโ๐ป that leverages the power of Retrieval Augmented Generation (RAG) to transform your learning experience. Unlike traditional online chatbots ๐ค, Dahih Al-Dofaa focuses on providing answers derived exclusively from your uploaded documents. By uploading textbooks, notes, and research papers, you create a personalized knowledge base accessible anytime, anywhere, without requiring an internet connection. ๐
Dahih prioritizes user privacy by keeping all data local ๐. Your documents and generated information remain securely on your device. With Dahih, you can study confidently, knowing your sensitive information is protected. ๐ช
Experience the future of learning with Dahih. Get accurate answers, deepen your understanding, and unlock your full academic potential. ๐
Dahih is available for download on Windows ๐ป, Debian-based Linux ๐ง, and RPM-based Linux systems ๐ฅ๏ธ. To get started, follow these simple steps:
Note: Ensure your system has at least 8GB of RAM. ๐ง
Upon launching Dahih for the first time, you'll be prompted to download and select an LLM (Large Language Model) ๐ง and an embedding model. These models are essential for the app's functionality. โก
Recommended Models:
Please note that these are suggestions based on common hardware configurations and model performance. You may explore and experiment with other models depending on your specific needs and system resources. ๐งช
Important: Ensure you have enough storage space to accommodate the downloaded models. ๐พ
The app will guide you through the download and installation process for the selected models.
Dahih is a revolutionary study tool designed to help students unlock their full potential. ๐ Unlike traditional online chatbots ๐ค, Dahih focuses on providing answers derived directly from your own study materials. ๐ By uploading textbooks, notes, and research papers, you create a personalized knowledge base accessible anytime, anywhere. ๐
Dahih is specifically designed for students who value privacy ๐ and efficiency. As a local application, it operates offline, ensuring your data remains secure and accessible without an internet connection. ๐ซ This makes Dahih the perfect companion for focused study sessions and environments with limited connectivity. ๐ถ
Our primary target audience includes university and college students seeking a more effective and private study solution. ๐งโ๐๐ฉโ๐
Dahih combines the power of AI with the convenience of offline access to provide an unparalleled study experience.
Dahih is built on a robust foundation of open-source technologies:
Dahih leverages the power of Ollama to support a diverse range of Large Language Models (LLMs), including:
This flexibility allows users to choose the model that best suits their needs and computational resources. Dahih's architecture is designed to accommodate future LLM integrations as they become available.
Prerequisites:
Ollama Executables:
Before running or building Dahih, you need to download and add the appropriate Ollama executables to the project directory. Ollama executables power the LLM functionality within the app.
ollama-linux-amd64
file.ollama-windows-amd64.zip
file, and extract its contents.extraResources/ollama
directory within the Dahih project.Environment Setup
For development purposes, you can customize the Ollama host URL by creating a .env
file at the root of the project and adding the following line:
OLLAMA_HOST=http://your_ollama_host:port
If this environment variable is not set, the app will default to http://localhost:11434
.
Installation:
git clone https://github.com/ibrahimhabibeg/Dahih-Al-Dofaa.git
cd Dahih-Al-Dofaa
yarn install
Development Setup:
To run the app in development mode:
yarn start
This will open the application in a new window.
Building the App:
To build the app for production:
yarn make
Note: This command builds executables specifically for the developer's operating system and architecture.
Additional Notes:
Dahih utilizes GitHub Actions for automated builds and releases. The build process is defined in the .github/workflows/build.yml file. This file outlines the steps involved in building the app for Windows and Linux platforms.
For detailed insights into the build process, please refer to the build.yml file. Understanding its contents can be invaluable for troubleshooting build-related issues.
By following these steps, you can successfully set up Dahih for development or build it for your system.
If you have any further questions or encounter issues, please don't hesitate to open an issue on the GitHub repository.
We welcome your feedback, suggestions, and contributions to Dahih. If you encounter any issues or have questions, please don't hesitate to reach out.
Preferred Contact Methods:
We strive to respond to all inquiries as promptly as possible.