Pricing
Usage-based pricing for the AI Gateway and Sandbox: no seats, minimums or monthly commitment, and optional plans set limits and top-up bonus. The enterprise platform is scoped and agreed in writing.
AI Gateway
One API for every model, routed to the cheapest available source, plus a flat 2.5% platform fee. Reach models from every major provider through one endpoint and one bill.
- Flat 2.5% platform fee on the lowest available rate
- Every major provider, one API key, one bill
- No hidden fees, no minimums, no commitment
Sandbox
Firecracker micro-VMs billed per second of vCPU and memory. No egress fees, and pause with your state preserved, so a paused sandbox is charged for memory and storage but not vCPU.
- Simple per-second rate by sandbox shape
- No egress fees
- Pause to stop vCPU billing when idle
Enterprise platform
Annual platform subscription, deployment tier, and an optional fixed-scope engagement. Scope, deliverables and cost agreed in writing before work starts.
- Platform scoped to the workflows and agents you run in production
- Deployment priced by where it runs: CreateOS cloud, your VPC, or on-premise
- Forward-deployed engineering for your first workflow
Need volume terms for the Gateway or Sandbox? Book a demo and we will model your usage with you.
Bring us one stuck workflow.
We take it to governed production with you, on your systems, in weeks, not quarters.