ReMind
Recall AI, also known as ReMind, is an AI tool designed to act as a local, artificial memory for your device. It captures digital activity and utilizes it as a ...
Last verified:
What is ReMind?
ReMind (also called Remind AI) is a local, privacy-focused artificial memory application that captures and indexes your digital activities on your device. The tool records screenshots, audio, and other digital interactions, then transcribes text from captured screenshots using OCR technology. It uses advanced AI models powered by Ollama with local LLMs like llama3.1 to provide detailed summaries of your daily activities and answer questions based on your digital history.
Key features include automatic capture of digital activities, text transcription from screenshots, vector database indexing for efficient retrieval, detailed daily activity summarization, and an interactive chat interface for querying your digital history. The application uses ChromaDB for vector storage and supports natural language queries about past activities. It's built with TypeScript (72.7%), Python (25.1%), and uses Electron for the desktop application.
ReMind is designed for productivity-focused users who want to boost their workflow by remembering tasks, meetings, documents, emails, and social media interactions. It's particularly useful for people who struggle with recalling digital information, professionals who need to track their daily work activities, and anyone wanting a personal artificial memory on their device. The tool is open-source under Apache-2.0 license, making it suitable for developers who want to customize or contribute to the project.
ReMind pricing
Pricing model: Free
Free - ReMind is 100% open-source under Apache-2.0 license and available for free. The macOS version is available for download, and the source code is freely accessible on GitHub at github.com/DonTizi/ReMind. There are no paid plans as the tool is completely free for personal usage and customization.
ReMind pros
- 100% open-source with Apache-2.0 license for full customization
- Local processing ensures all data stays private on your device
- No external cloud dependencies - complete data security
- Integrates with Ollama for secure on-device LLM processing
- Captures screenshots, audio, and digital activities automatically
- OCR text transcription from captured screenshots
- Vector database indexing with ChromaDB for fast retrieval
- Provides detailed daily activity summaries
- Interactive chat interface to query digital history
- Natural language queries for accessing past activities
- macOS version available for download with beautiful UI
- UI powered by Enchanted from AugustDev
- Efficient indexing retrieves documents quickly
- Fully customizable to fit individual needs
- Community-driven development with active contributor base
- No third-party cloud software integration required
- Local encryption keeps data accessible only to you
ReMind cons
- Requires manual setup with Node.js 18+, npm, Ollama, and Python
- First use requires waiting 30 minutes to gather activities into database
- Only macOS version officially available for download
- No pre-built releases published on GitHub yet
- Requires pulling specific llama3.1 model for Ollama
- Electron-based app may have higher memory consumption
- Limited to local LLMs - no cloud AI model options
- Active development status means potential stability issues
- Setup process requires technical knowledge for non-developers
- No Windows or Linux version currently available
Frequently asked questions about ReMind
What is ReMind AI?
ReMind AI is a local, privacy-focused artificial memory application that captures and indexes your digital activities on your device. It transcribes and summarizes your digital interactions for easy recall, using advanced AI models powered by Ollama with local LLMs to provide detailed summaries of daily activities and answer questions based on your digital history.
How do I set up ReMind?
To set up ReMind: 1) Clone the repository using git clone https://github.com/DonTizi/ReMind/, 2) Install Node.js 18+ and npm, 3) Download Ollama and pull llama3.1, 4) Download Python, 5) Navigate to the project directory and run npm install, 6) Start the server with npm run dev. For macOS users, a first version is available for direct download.
Is ReMind private and secure?
Yes, ReMind is privacy-focused by design. It uses local processing with Ollama for AI, ensuring all data stays on your device. There are no external cloud dependencies or third-party software integrations. Your data is locally encrypted and stored, only accessible by you, with no transmission to external servers.
What digital activities does ReMind capture?
ReMind captures screenshots, audio, and other digital activities including emails, documents, social media posts, and web interactions. It uses OCR technology to transcribe text from captured screenshots, creating a comprehensive index of your digital footprint.
What AI models does ReMind use?
ReMind uses Ollama for local LLM processing and requires you to download and pull the llama3.1 model. The application integrates with local language models to provide summarization and answer questions about your digital history without sending data to external AI services.
Can I customize ReMind?
Yes, ReMind is 100% open-source under Apache-2.0 license, making it fully customizable to fit your needs. You can access the source code on GitHub, modify the TypeScript and Python codebase, and contribute to development. The open-source nature allows complete control over data handling and customization.
How do I query my digital history with ReMind?
After launching the ReMind application, you can interact through a chat interface to ask questions about your digital activity. The tool supports natural language queries, allowing you to search for past activities, meetings, documents, or any digital interaction using everyday language rather than specific keywords.
What platforms does ReMind support?
ReMind currently has a macOS version available for download with a beautiful UI powered by Enchanted. The source code is available for Linux and Windows, but only the macOS version is officially released. The application is built with Electron, TypeScript, and Python.
How long does it take for ReMind to start working?
On first use, you should wait about 30 minutes to allow the application to gather digital activities into its vector database (ChromaDB). After this initial period, ReMind will have collected enough data to provide summaries and answer questions about your digital history effectively.
Can I contribute to ReMind development?
Yes, ReMind welcomes community contributions. To contribute: 1) Fork the repository, 2) Create a new branch with git checkout -b feature-branch, 3) Make your changes, 4) Commit with git commit -m 'Add new feature', 5) Push to the branch, and 6) Create a pull request. The project has 485 stars and 3 contributors on GitHub.