SurfSense
Open-source NotebookLM / Perplexity-style research workspace you can self-host.
Quick answer
SurfSense is an open-source NotebookLM / Perplexity-style research workspace you can self-host, listed under AI search & research assistants. A typical small-business use: team research workspace that answers questions over uploaded docs and connected sources. The project publishes no official RAM figure, so we do not estimate a monthly server cost. License: Apache-2.0, which allows internal business use.
Last updated · facts verified against official docs on · sources
Activity data comes from GitHub and was last refreshed Oct 25, 2026.
What is SurfSense?
SurfSense is a self-hostable research and note-taking workspace with a web UI, connectors, document parsing and chat over your own sources. The Docker stack ships PostgreSQL with pgvector, Redis, Celery workers, a FastAPI backend, frontend and a Caddy reverse proxy on one origin. It targets an air-gapped alternative to NotebookLM and Perplexity.
Where it fits in a 10–200 person business
- Team research workspace that answers questions over uploaded docs and connected sources
- Self-hosted NotebookLM-style notebook with your own LLM and embedding endpoints
- Shared research spaces for product, legal or support teams that cannot use cloud AI notebooks
These use cases are our editorial suggestions, not claims made by the project.
Can a business use SurfSense? (license)
| License | Apache-2.0 |
|---|---|
| Type | Permissive: MIT / Apache-2.0: use, modify and sell with few obligations. |
| Internal business use | Allowed under the license text |
| What to watch | Permissive. Internal and commercial use allowed; keep the Apache-2.0 license and notices. |
This is a plain-English summary of the license text, not legal advice. Third-party models you run with it have their own licenses.
Data & compliance: does SurfSense phone home?
| Telemetry (usage data sent to the vendor) | Off by default (opt-in) ADR 0016 (Accepted, 2026-09-08) states the app sends no telemetry or crash reports and that declared dependencies include no analytics SDK. The ADR is part of the local/desktop pivot docs; the Docker stack is not called out separately. How to turn it off: Not needed per ADR 0016. Source: github.com/MODSetter/SurfSense/blob/main/docs/adr/0016-no-telemetry.md |
|---|---|
| Other calls home by default (updates, marketplaces, default cloud services) | Not documented Related ADR policy is egress-off-by-default with user consent for outbound calls. Any LLM, embedding, connector or parsing APIs you configure will make outbound calls. How to turn it off: Leave external providers unset; use local Docling/sentence-transformers defaults. |
| Single sign-on (OIDC / SAML / LDAP) | Not documented Auth documented as LOCAL (email/password) or Google OAuth. No SAML/OIDC/LDAP page found. |
| Audit log | Not documented |
| Role-based access control (RBAC) | Not documented |
| Data residency / GDPR notes from the project | Not documented |
Checked against the project's own docs and repository on . "Not documented" means we could not find the answer in official sources; it does not mean the feature is missing or that no data is sent. Third-party services you connect (model APIs, vector databases, search engines) have their own data policies. Compare all apps.
What hardware does SurfSense need?
The project does not publish a minimum CPU/RAM figure, so we leave the official value blank rather than guess.
GPU: Optional. Set SURFSENSE_VARIANT=cuda (or cuda126) and the GPU Compose overlay when you want NVIDIA acceleration; CPU images are the default.
No official min/recommended CPU or RAM figures are published. Prerequisites are Docker and Docker Compose. GPU needs the NVIDIA Container Toolkit on the host.
How much does it cost to host SurfSense per month?
The project publishes no RAM figure we could verify, so we do not estimate a cost.
How to deploy SurfSense
Method: Docker Compose
# Option 1 — one-line install script (recommended in the docs)
curl -fsSL https://raw.githubusercontent.com/MODSetter/SurfSense/main/docker/scripts/install.sh | bash
# Option 2 — manual
# git clone https://github.com/MODSetter/SurfSense.git
# cd SurfSense/docker
# cp .env.example .env # set SECRET_KEY (openssl rand -base64 32)
# docker compose up -d
# UI: http://localhost:3929From the official instructions: www.surfsense.com/docs/docker-installation. Always check the source for the current version before you run it.
Known gotchas when self-hosting SurfSense
- SECRET_KEY is required in .env (generate with openssl rand -base64 32). Leaving the placeholder is unsafe on any shared host. source
- The install script may offer Watchtower for daily image updates; those pulls can be several GB. Pass --no-watchtower on the script if you want to skip it. source
- Docs still label some Docker paths under /docs/legacy/ while calling Docker the recommended install; follow the live Docker Installation page and pin SURFSENSE_VERSION instead of latest for production. source
- Default embedding uses a local sentence-transformers model; switching to OpenAI/Cohere/Ollama or Unstructured/LlamaCloud parsing sends data to those providers once configured. source
Alternatives in AI search & research assistants
- Khoj Personal and team AI "second brain" for your documents and the web. Compare Khoj vs SurfSense
- Vane (formerly Perplexica) Private, Perplexity-style AI answer engine with cited web sources. Compare SurfSense vs Vane (formerly Perplexica)
- GPT Researcher Open-source deep research agent that browses the web and local documents and writes cited reports. Compare GPT Researcher vs SurfSense
SurfSense FAQ
How much RAM does SurfSense need?
SurfSense does not publish an official RAM figure, so we do not state one. Prerequisites are Docker and Docker Compose. GPU needs the NVIDIA Container Toolkit on the host.
Does SurfSense need a GPU?
Optional. Set SURFSENSE_VARIANT=cuda (or cuda126) and the GPU Compose overlay when you want NVIDIA acceleration; CPU images are the default.
How much does it cost to self-host SurfSense?
The software is free to download. The project publishes no official RAM figure, so we do not estimate a monthly server cost.
Is SurfSense free for commercial use?
SurfSense is licensed under Apache-2.0 (Permissive). Internal business use is allowed under the license text. Permissive. Internal and commercial use allowed; keep the Apache-2.0 license and notices. This is a summary, not legal advice.
What are common problems when self-hosting SurfSense?
SECRET_KEY is required in .env (generate with openssl rand -base64 32). Leaving the placeholder is unsafe on any shared host. The install script may offer Watchtower for daily image updates; those pulls can be several GB. Pass --no-watchtower on the script if you want to skip it. Docs still label some Docker paths under /docs/legacy/ while calling Docker the recommended install; follow the live Docker Installation page and pin SURFSENSE_VERSION instead of latest for production. Default embedding uses a local sentence-transformers model; switching to OpenAI/Cohere/Ollama or Unstructured/LlamaCloud parsing sends data to those providers once configured.
Is SurfSense still maintained?
Yes, it is actively developed. Last commit: Oct 21, 2026. Latest release: v2.2.3 (Oct 21, 2026). Checked Oct 25, 2026.
How do I install SurfSense?
The official method is Docker Compose. The exact commands are in the "How to deploy" section above and in the official instructions: https://www.surfsense.com/docs/docker-installation
What are self-hosted alternatives to SurfSense?
Other apps in AI search & research assistants: Khoj, Vane (formerly Perplexica), GPT Researcher.
Sources
- GitHub repository: github.com/MODSetter/SurfSense
- Official documentation, Hardware requirements, Deploy instructions, Gotcha 1, Gotcha 2, Gotcha 3, Gotcha 4: www.surfsense.com/docs/docker-installation
- License text: github.com/MODSetter/SurfSense/blob/main/LICENSE
- Hetzner Cloud pricing (cost estimate): www.hetzner.com/cloud/regular-performance
- GitHub stars, commits and releases: GitHub REST API, fetched
Facts on this page were checked against the project's own repository and docs on 2026-10-25. Spotted something out of date? Tell us.