SafeRag is your private AI assistant powered by local LLMs. All processing happens on your Mac—no cloud, no data exposure, complete privacy. GDPR and HIPAA compliant out of the box.
Requires macOS 14.0 or later • Built with SwiftUI • Powered by Ollama
hero-saferag-chat.png
All AI processing happens on your Mac using Ollama. No data ever leaves your device. No cloud connections, no tracking, complete control.
Built-in compliance features: complete audit logs, data export, right to deletion, encryption status reporting. Enterprise-ready from day one.
Upload your documents (PDF, DOCX, TXT, MD) and chat with them. Advanced vector search and context-aware responses powered by local embeddings.
Built entirely with SwiftUI. Fast, efficient, beautiful. Feels like Apple made it. Menu bar integration, keyboard shortcuts, and system-native design.
Support multiple users with role-based access control. Admin console for user management. Perfect for teams and small organizations.
Point SafeRag at folders and automatically process new documents as they arrive. Zero-friction knowledge base building.
gallery-chat-rag.png
gallery-documents.png
gallery-compliance.png
gallery-models.png
Download SafeRag for macOS. The app automatically installs Ollama if needed and helps you download your first AI model.
Set up your local account (first user gets admin privileges). Everything runs locally on your Mac—no cloud registration required.
Upload documents for RAG intelligence, choose your AI model, and start having private conversations that never leave your Mac.
Discuss patient cases with AI while maintaining HIPAA compliance. No PHI ever leaves your Mac.
Analyze legal documents and develop case strategies without violating attorney-client privilege.
Use AI for proprietary information, trade secrets, and confidential business planning without cloud exposure.
Get AI assistance without surrendering your data to big tech. True privacy by design.
Join privacy-conscious professionals who refuse to compromise on data security.
Download NowVersion 1.0 • macOS 14.0+
SafeRag runs entirely on your Mac using local AI models (via Ollama). Unlike cloud-based services, your data never leaves your device. You get AI capabilities without privacy compromises.
SafeRag includes one-click Ollama installation and model management. The app guides you through downloading your first AI model. Everything is automated.
SafeRag supports all Ollama models including Llama 3.2, Mistral, Gemma, and many more. You can download and switch between models instantly within the app.
Yes. Since all data stays on your Mac, there's no data processor or third-party risk. SafeRag includes audit logs, data export, and encryption reporting required for compliance.
Upload documents and SafeRag creates local vector embeddings. When you ask questions, the app finds relevant document sections and includes them as context for AI responses.
Yes! SafeRag supports multiple users with individual accounts. The first user becomes admin and can create additional user accounts with different permission levels.
SafeRag requires macOS 14.0 or later. For best performance, we recommend Apple Silicon (M1/M2/M3) with at least 8GB RAM. Works on Intel Macs but may be slower.
Currently SafeRag is macOS-only. An iOS companion app for viewing chat history is on our roadmap.
SafeRag is built with the help of these excellent open source projects. We're grateful to their maintainers and contributors.