Frequently asked questions

Ten questions that come up most. If yours isn't here, open a thread on GitHub Discussions — we'd rather hear it once than have ten people stub their toes.

• +−Which LLM does OpenMind use?

  By default Anthropic's Claude Sonnet 4.6, picked for the extraction quality on conversational German + English. Power users can switch per project to OpenAI, Ollama (local), or LM Studio (local) via Settings → LLM provider. Bring your own API key in Settings → API keys; without one the hosted demo uses the maintainers' keys with a per-user soft cap.
• +−Where does my data live?

  On the hosted demo: a Supabase Postgres database in EU-central, row-level-security-locked to your account. On self-host: wherever you put it (your VPS, your laptop, your air-gapped box). See /help/privacy for the full breakdown — what we store, who can read it, how to export everything, how to delete your account.
• +−Can I bring my own API key?

  Yes — Settings → API keys → Anthropic / OpenAI. Stored per-user (RLS-scoped). Once set, every extraction and Ask-KG call from your account uses your key, billed directly by the provider. No proxy, no inflation.
• +−What attachments can I drop into the chat?

  TXT, MD, PDF, DOCX are extracted in-process. Audio and video (MP4, MOV, MKV, WebM, MP3, WAV) are queued to a worker that runs faster-whisper for transcription — available on self-host; the hosted demo rejects video uploads with HTTP 422 to keep the free tier sustainable. Drag-drop or paperclip into the composer.
• +−Why is a node tagged [uncertain]?

  The extractor returns a confidence per node. Anything below 0.6 lands with the [uncertain] tag — usually a sign the model wasn't sure how to canonicalise an entity (an ambiguous pronoun, a tentative decision, a half-described risk). Click the node, refine the wording in chat, and it usually firms up on the next pass.
• +−Can I share a project with someone else?

  Yes — kebab on a project card → Transfer ownership. Currently the model is single-owner; the new owner gets full control and you stay as an editor. A multi-member share without ownership transfer is on the roadmap (workstream-ui-completeness).
• +−How do exports work?

  Every project has an "Export ▾" dropdown in the workspace header with six formats: Markdown, Mermaid, OPML, FreeMind, PNG, SVG. /help/exports breaks down when to use which. Mermaid is canonical — PNG and SVG render from it, so what you see in Mermaid is what you get as an image.
• +−Is there an API I can call?

  Yes — `/api/public/v1/...`, gated by personal access tokens you mint in Settings → API keys. Read-only today (projects, conversations, messages, nodes, edges). The full schema is documented at https://api.openmind.app/docs (FastAPI's auto-generated OpenAPI page).
• +−Can I run OpenMind offline / self-hosted?

  Yes — Apache-2.0, full docker-compose stack at the repo root. `docker compose up -d` on a $5/month VPS gives you a working instance: web + API + worker + Postgres + render sidecar. Pair with Ollama on the same host for end-to-end offline. See docs/self-hosting.md.
• +−I clicked Delete account — is it really gone?

  Yes. The destructive action cascades immediately: auth row → profiles → projects → conversations → messages → nodes → edges → attachments → embeddings — every dependent row drops on the same SQL transaction. There is no soft-delete grace period. Type-the-email-to-confirm gates the action; once submitted, it's irreversible. Use Settings → Account → Download my data first if you want a backup.
• +−If I paste a whole article, will OpenMind read all of it?

  Yes. Long pastes (over ~2 000 tokens) are split into segments behind the scenes; the extractor walks every segment in sequence, threading earlier-segment nodes into the next so the same entity isn't duplicated across them. The success badge under your message shows 'from N segments' once it's done so you can confirm everything was processed. Watch the project's training-corpus stats grow at /settings#training.
• +−Can I edit or delete a node directly?

  Delete: yes — search for the node in the mindmap pane, hover the result row, click the trash icon. Edges and attributes touching the node disappear automatically (FK cascade). Re-mention the corrected concept in chat and the LLM creates a fresh node. Recategorize (change the kind): yes — multi-select rows in the search panel and pick a new kind from the toolbar. The label, summary, and edges stay intact; only the bin changes, so the chat history and graph stay consistent. Edit (label / summary / confidence): no end-user affordance, intentionally — it would drift the graph from the chat history and pollute the training-corpus signal. See ADR 0010 for the trade-offs.

Still stuck? GitHub Discussions is the fastest way to reach a human.