Getting started
Sign in, create your first project, talk through it, watch the mindmap form. Five minutes.
Help
Back to the appOpenMind help
Short, on-brand answers to the questions people actually ask. If something here is wrong or missing, open an issue on GitHub — we'd rather hear it once than have ten people stub their toes on it.
Start here
Start here Architecture How it works
Talk it. Map it. Own it. — the three landing pillars unpacked. What the LLM extractor actually does, what RLS gives you, what self-host means in practice.
FAQ Frequently asked questions
Ten questions that come up most — providers, attachments, exports, sharing, self-host, account deletion.
Using OpenMind
Mode Ask-KG mode
Read from the graph instead of feeding it. When to use it, what [uncertain] means, and how source-node attribution works.
Files Attachments
What happens when you drop a PDF, DOCX, audio, or video into chat. Caps, extraction times, hosted vs self-host.
Export Pick the right format
Markdown, Mermaid, OPML, FreeMind, PNG, SVG — same data, different shape. Quick guide for which to use when.
LLM Pick a provider
Anthropic, OpenAI, Ollama, LM Studio — when each makes sense + how to bring your own key.
Own it
Own it Privacy & your data
Where your conversations live, what we keep, how to export everything, and how to delete your account.
Build Public API
Read-only REST at /api/public/v1, token-authenticated, RLS-scoped. Mint a token, curl an endpoint.
Run it Self-host
Five steps from git clone to working instance on a $5 VPS. End-to-end offline path with Ollama.
Train Training corpus
Phase-4 SLM workflow — opt projects in, watch counts on /settings, run the CLI when ready. Bridge between dashboard and CLI.
Source Open source
OpenMind is Apache-2.0 — read the code, fork it, ship a self-host. Star the repo if you like what you see.