Open Notebook

Open Notebook is an alternative to Google's NotebookLM for researchers, students and professionals working with their own materials. It brings research and notes together while letting you choose what context an AI model can access.

• Research collections: Organize PDFs, Office documents, web pages, video and audio into separate notebooks alongside notes you write yourself.
• Cited conversations: Ask questions about your research and get answers that reference the sources, notes and insights they used.
• Search and summaries: Find matching words with full-text search or related content with vector search, and use customizable AI actions to summarize material or extract ideas.
• Podcast generation: Turn notes into audio discussions with customizable voices, speakers and episode profiles.

The app is open source under the MIT license and runs locally or on your server through Docker on macOS, Windows and Linux. Local model connections include Ollama, LM Studio and other OpenAI-compatible endpoints. It also supports oMLX on Apple Silicon and reasoning models such as DeepSeek-R1 and Qwen3.

You can connect cloud providers including OpenAI, Anthropic and Google. Those providers receive the context used in their requests, so self-hosting alone doesn't keep every AI interaction local. Speech recognition and voice generation support depend on the provider you choose. A REST API provides access for custom integrations, and optional password protection is available for public deployments.