Pricing

CleverRouter is pay-as-you-go on top of the underlying provider cost. The gateway adds a flat 5% markup. No subscription, no per-seat fees during the Beta.

At a glance


Markup	Provider cost + flat 5% on actual token usage
Beta tier	€0 + 5%, 60 RPM per key
Per-key budget	Caps daily/monthly spend, returns `402`
Failed calls	Free (`4xx` and `5xx` don't bill)

How billing works

Provider cost + 5% markup on actual token usage. We charge what the provider charges us, plus a flat 5% to cover the gateway.
No subscription, no monthly minimum, no per-seat pricing during Beta.
Per-key budgets in the dashboard cap spend before it happens — hitting a budget returns 402 budget_exceeded (CRBudgetExceededError).

Beta tier (current default)

Limit	Value
Price	€0 + 5% on provider costs
Rate limit	60 RPM per key
Organisations	1
Team members	up to 3
Logs retention	30 days (metadata only)
Models	All chat, embeddings, audio, rerank
Zero data retention	Enabled by default
Support	Community Slack

Every Beta workspace starts with €5 of credit. See cleverouter.eu/pricing for the live comparison.

Pro (coming soon)

Limit	Value
Price	€29/month + 5% on provider costs
Rate limit	600 RPM per key
Organisations	up to 3
Team members	up to 25
Logs retention	180 days
SSO	Google Workspace, Microsoft Entra
Support	Priority email + provider-incident alerts

Enterprise

Custom pricing with volume discounts on the provider markup. For teams above 1k RPM, procurement-led with regulatory commitments. Email hello@clevermation.com.

What we never charge for

Listing models (GET /v1/models).
Provider health checks.
Failed requests (4xx and 5xx).

Monitoring spend

Dashboard → Usage shows daily aggregates per model, per key, per day.
The requests table records metadata only (tokens, latency, status, cost in cents). No prompts, no completions — see Zero data retention.
Usage is aggregated every 5 minutes from requests → usage_daily by the gateway's rollup job.

Token prices vary per model

Token prices vary per model and provider. See the live catalog at cleverouter.eu/models for current input/output rates per 1M tokens.

Per-key budgets

// API: not directly callable; budgets are set in the dashboard.
// Behaviour at runtime:
import { CRBudgetExceededError } from '@cleverrouter/sdk';

try {
  await cr.chat({ model, messages });
} catch (err) {
  if (err instanceof CRBudgetExceededError) {
    // 402 — top up or wait until tomorrow / next month
  } else {
    throw err;
  }
}

Budgets are checked against usage_daily aggregates at request time.

Rate limits — RPM enforcement.
Get your API key — set per-key budgets.
Errors — CRBudgetExceededError.

On this page