Settings

Your keys and preferences live only in this browser. Nothing here is sent to us — clearing your browser data removes it. Use Export to carry your profile to another device.

Bring your own AI key. Keys are stored only in this browser (IndexedDB); we never store or log them. Groq and NVIDIA give free API keys, so the AI tools can run at no cost. Note: whichever provider you pick, your prompt text is sent to that provider's servers (through our no-log proxy) and processed under their terms — the AI tools are not on-device. For fully local work, use the offline tools.

Groq (free, fast)key from console.groq.com/keys

🎁 Free: sign in at console.groq.com and create an API key — no card needed.

NVIDIA NIM (free)key from build.nvidia.com

🎁 Free: sign in at build.nvidia.com, pick a model, and click “Get API Key”.

Anthropic (Claude)key from console.anthropic.com
OpenAI (GPT)key from platform.openai.com
xAI (Grok)key from console.x.ai
Custom / self-hosted (your own cloud)key from your own OpenAI-compatible endpoint · stored for your own use

🎁 Fully private: run Ollama (ollama.com), LM Studio, vLLM or LocalAI on your machine or cloud, then set its base URL (e.g. http://localhost:11434/v1). Your text goes only to your endpoint — never to us or any third party.

Bring your own cloud

For maximum privacy, point the AI tools at your own model (the “Custom / self-hosted” option above — Ollama, LM Studio, vLLM), and run the heavy/native tools in your own runner below. In both cases data goes only to your endpoint — nothing reaches Demystify or any third party.

Your tools runner (Docker)for heavy tools (office, PDF/A, video encode) — runs in your own cloud

🐳 Optional. Run docker run -p 8787:8787 demystify/tools-runner on your machine or cloud, then paste its URL. Heavy tools send data only to your runner — never to us.

My modules

Pick the product lines you use to focus the launcher. Leave all unchecked to see every tool. Stored only in this browser.