The provider’s price,
plus 4.5%.
We add 4.5% on top of the provider's own token price. That underlying price is never marked up, and every model shows what it costs in the dashboard before you spend anything.
// How billing works
Four sentences, and then the invoice makes sense.
Prepaid credits, one invoice, and a spend limit per key. There are no seats to licence, no minimum on self-serve, and the total at the end of the month is the total you watched accumulate.
- 01
The provider's price, unchanged
Whatever the upstream charges for a million tokens is what flows through. We do not mark it up, and we do not quietly route you to a worse-value endpoint to widen a margin.
- 02
A visible 4.5% on top
One line, one number, on the routed spend. Self-serve gateways typically charge 5% to 5.5%; in-region routing usually sits behind their business tiers on top of that.
- 03
Live prices before you spend
Every model in the catalog shows its current price and speed per route in the dashboard. The quoted price is the price that settles, whoever ends up serving the call.
- 04
Discounts at volume
Committed-use accounts get an enterprise discount on routed spend, along with invoicing, an SLA and a named contact. Pricing is scoped to your models and volumes.
Where are the numbers?
Per-model prices move with the market, so the table lives on the price list, read live from the catalog: text, image, video, speech and music, one published price per model. No account needed to read it.
We took the comparison figures from other gateways’ own pricing pages in September 2026. Enterprise discounts, EU-only routing and private endpoints are quoted per account. Write to [email protected].
Our fee is one line
on the invoice.
We add 4.5% on top of the provider’s own token price, and that price is never marked up. Self-serve gateways typically charge 5% to 5.5%. Every model’s live price sits in the dashboard before you spend a cent.
- Create an account
Self-serve
Prepaid credits
Create an account, mint a key, top it up. Every model shows its live price in the dashboard before you spend anything.
- Live per-model prices and speed
- Spend limit, expiry and model allowlist per key
- Usage analytics and spend alerts
- All five modalities on one key
- Best value at volumeContact sales
Committed use
Annual agreement
Volume discounts on pooled spend, EU or global routing, invoicing and a named contact. Priced to your models and volumes.
- Enterprise discount on routed spend
- EU or global routing set per key
- Invoicing, SLA and dedicated support
- Custom fallback policies, audit logs, SSO
- Talk to us
Resellers
Credit batches
Telcos, ISPs, integrators and platforms bundling prepaid AI credits for their own customers, with a margin on top.
- Prepaid credit batches at a discount
- Sub-keys with their own budgets
- Per-project usage reporting
- White-glove onboarding
Enterprise discounts, EU-only routing and private endpoints are quoted per account. Write to [email protected] or use the form on the contact page.
The questions we
actually get asked.
You keep one key, one invoice and one integration instead of ten. On every call we compare price, latency and availability across every route that can serve the model you asked for, and dispatch to the one your policy prefers. When a route fails before it has sent anything, the request moves to another endpoint serving the same model.