Memorwise
An open-source, local-first alternative to NotebookLM that lets a person chat with their own PDFs, images, audio, video, URLs and YouTube links using whichever LLM provider they choose, including fully local models.
Memorwise installs and runs with a single npx command, which clones the project, installs dependencies, starts a local server and opens a browser to it; from there a user connects at least one of eight supported LLM providers, including Ollama, LM Studio, OpenAI, Anthropic, Gemini, Groq, Mistral and OpenRouter, creates a notebook, adds sources and starts chatting. Documents and other sources are chunked and embedded on the user's own machine before any chat happens.
Beyond retrieval-augmented question answering with source citations, it generates flashcards, quizzes, study guides and summaries per notebook, extracts a knowledge graph of AI-identified concepts showing how sources connect, and can generate an audio overview presented as a podcast-style multi-speaker discussion of the source material. It supports more than 20 file formats, including PDF, DOCX, XLSX, images through OCR, and audio or video through Whisper transcription.
It is written in TypeScript with Next.js, requires Node.js 22.13 or newer, and is released under the MIT license.