Simple, transparent pricing

Start free with local models. Connect to the relay when you're ready. Pro unlocks near-wholesale rates.

Monthly
Annual (2 months off)

Free

$0
forever, no sign-in required

A complete AI workspace that runs on your machine. Bring your own keys.

  • Local AI via llama.cpp
  • BYOK direct to every provider
  • Full memory · knowledge graph · profiles · RAG · pipelines · browser tool
  • Sign-in-with-Claude (use your Anthropic subscription)
  • Up to 5 projects
  • Up to 5 active automations
  • No relay access
  • No device sync
  • No desktop tunnel
Download
Most flexible

Pay-as-you-go

1.3×
over wholesale, no monthly fee

Sign in. One bill across every provider. Pay only for what you use.

  • Everything in Free
  • 5 projects & 5 active automations (Pro lifts both)
  • One bill across every LLM provider
  • Anonymous routing — providers see Central-Intel, not you
  • Device sync (up to 10), E2E encrypted
  • Desktop tunnel
  • $1 of free credits to start
Sign in

Pro

$15/mo
1.02× wholesale on all usage

For people who use AI every day. Near-wholesale rates plus everything we build.

  • Everything in Pay-as-you-go
  • 1.02× wholesale — save 22% vs PAYG on every call
  • Unlimited projects + unlimited automations
  • Relay-brokered integrations as they ship (Calendar, etc.)
  • Mobile app + tunnel when those ship
  • Priority support
Become a member

How pricing works

Central-Intel routes your LLM requests through our relay so providers never see your identity. Pay-as-you-go users pay 1.3× the provider's wholesale rate per call. Pro members pay 1.02× — barely above wholesale — and unlock unlimited projects, unlimited automations, and the platform integrations as we ship them.

There's no monthly minimum, no expiring credits, and no contract. Every transaction shows the base provider cost and the service fee side-by-side so you always know what you're paying for.

Payment processing fees apply at checkout. US-only at launch.

Have questions? Check our Frequently Asked Questions or contact us.