Memfect

Memfect transforms scattered research notes into an interconnected knowledge graph. Key features include: • Ingests papers, PDFs, YouTube videos, and Git repositories • Generates atomic notes and links them automatically • Semantic search across your personal knowledge base • Local-first operation with no account or cloud dependency This application is designed to prevent valuable research from being lost in disorganized folders. It takes diverse source materials and processes them on your local machine, building a living, interconnected knowledge atlas. Users can drop in a source file, and the application immediately creates concise, individual notes, linking them logically within the existing knowledge graph. The system offers powerful semantic search capabilities, allowing users to query their own collected knowledge rather than relying on external web searches. This ensures that answers are derived directly from the user's specific research and learning. Memfect supports various large language models, including OpenAI, Gemini, Claude, and local options via Ollama, giving users flexibility in processing. Memfect is ideal for researchers, academics, students, and anyone building a personal knowledge base who values data privacy and offline access. It helps individuals in fields like physics, chemistry, and software engineering to quickly synthesize information, discover connections, and maintain a robust, searchable repository of their insights.