SurfSense: what it is, what problem it solves & why it's gaining traction

What it solves

SurfSense is an open-source alternative to Google NotebookLM, designed to remove limits on the number of sources and notebooks, eliminate vendor lock-in, and provide greater control over data privacy. It addresses the constraints of proprietary AI notebooks by allowing users to configure their own LLMs and embedding models while integrating with a vast array of external data sources.

How it works

SurfSense creates a searchable knowledge base from synced external connectors (such as Notion, Slack, and Google Drive) and uploaded files. It uses a hybrid search (semantic and full-text) with hierarchical indices to provide cited answers. The system is powered by an agentic architecture using LangChain Deep Agents, enabling planning, sub-agents, and file system access. Users can interact with their data via a web interface, a desktop app for system-wide AI assistance, or a self-hosted Docker deployment.

Who it’s for

It is designed for researchers, teams, and individuals who need a comprehensive, private, and collaborative AI-powered knowledge management system that integrates with their existing toolstack and supports self-hosting.

Highlights

• Unlimited Sources: No limits on the number of notebooks or the size of individual sources.
• Model Flexibility: Support for 100+ LLMs via OpenAI spec and LiteLLM, including local models via vLLM and Ollama.
• Larga Integration: 25+ external connectors including Notion, Slack, GitHub, and Jira.
• AI Automations: Scheduled workflows, event-triggered agents, and no-code automations built via chat.
• Deliverable Studio: Tools to generate cited research reports, AI podcasts, editable presentations, and videos.
• Real-time Collaboration: Multiplayer support with RBAC and real-time chat/comments.
• Desktop App: System-wide assistance including screenshot capture and local folder syncing (e.g., for Obsidian vaults).