One command to turn any VPS, Mac Mini, or Raspberry Pi into a private AI node accessible from the TNode iOS app.
curl -fsSL https://install.tbrain.app | bashNo flags required. The installer auto-detects your hardware and configures everything (LLM runtime, gateway, Cloudflare Tunnel, pairing, telemetry).
| Hardware | Minimum |
|---|---|
| RAM | 1 GB (VPS) · 8 GB recommended for local LLM |
| Disk | 5 GB free |
| OS | Ubuntu 22.04+ / Debian 12+ / macOS 13+ / Raspberry Pi OS Bookworm |
| Network | Outbound HTTPS to Cloudflare, Firebase, and your LLM provider |
GPU is optional. If detected, the installer provisions a local model via Ollama; otherwise it falls back to a cloud provider over HTTPS.
# Default (auto-detect, Cloudflare Tunnel, no LAN exposure)
curl -fsSL https://install.tbrain.app | bash
# Bring your own OpenRouter key
curl -fsSL https://install.tbrain.app | bash -s -- --api openrouter --api-key sk-or-v1-...
# Local-network only (no tunnel; pair over Tailscale)
curl -fsSL https://install.tbrain.app | bash -s -- --no-tunnel --with-tailscalePass --help for the full flag list.
At the end of the install a QR code is printed. Scan it with the TNode app to complete device pairing.
- App: tbrain.app
- Issues with the installer: open one in this repo.
Proprietary — TBrain Platform. All rights reserved.