# Pricing

Celeris uses prepaid, per-token billing. Buy credits under
[**Settings → Billing**](https://console.celeris.ai/settings/billing), reached
from the account menu in the bottom-left of the [Console](/console#billing);
completed requests draw from the workspace's balance according to token usage.
There are no subscriptions, seat charges, or minimum commitments.

## Rates

| Tier | Prompt tokens | Completion tokens |
| --- | --- | --- |
| General | US$2 per million | US$6 per million |

## Estimating costs

Completed responses report exact token counts in their `usage` block:

```json
"usage": {"prompt_tokens": 27, "completion_tokens": 2, "total_tokens": 29}
```

Cost per call is `prompt_tokens × $2/M + completion_tokens × $6/M`. The
following examples show approximate costs for common
[agentic workloads](/agentic-workloads):

| Workload | Prompt | Completion | Cost per call | Calls per US$1 |
| --- | --- | --- | --- | --- |
| Classification / routing | ~30 tokens | ~4 tokens | ~$0.000084 | ~12,000 |
| Judge / rerank scoring | ~120 tokens | ~3 tokens | ~$0.000258 | ~3,900 |
| Structured extraction | ~200 tokens | ~48 tokens | ~$0.000688 | ~1,500 |
| Query rewriting | ~60 tokens | ~24 tokens | ~$0.000264 | ~3,800 |

Two request settings have the greatest effect on spend:

- **Set `max_tokens` deliberately.** Completion tokens cost three times as
  much as prompt tokens, and a limit bounds the cost of unexpected output.
  Supported values are positive multiples of 256 that fit the context window.
- **Keep prompts concise.** Repeated instruction tokens are billed on every
  call.

## What is — and isn't — charged

- **Served requests are charged** from their exact `usage` token counts, the
  same token volumes the [usage dashboard](/console#usage) aggregates.
- **Rejected requests are not charged.** This includes throttled
  ([`429`](/errors#429)), unauthorized ([`401`](/errors#401)), and
  out-of-credit ([`402`](/errors#402)) requests.
- **Server-side failures are free.** You are never charged for a
  [`502`](/errors#502) or [`503`](/errors#503).
- **A client disconnect does not guarantee cancellation.** If processing
  continues after a disconnect, the request may still be charged for work
  performed.

## Buying credits

Buy credits with a card (Stripe) under
[**Settings → Billing**](https://console.celeris.ai/settings/billing), reached
from the account menu in the bottom-left of the [Console](/console#billing), in
preset amounts of $5 / $10 / $50 / $100 or a custom amount between $5 and
$1,000 per purchase. You can buy credits with a payment card issued in a
country where Celeris is available (currently the United States).
See [Availability](/availability) if you're somewhere else. Receipts are
emailed automatically, and invoice documents remain available under
[**Settings → Billing**](https://console.celeris.ai/settings/billing).

When your balance reaches zero, API calls return
[`402 insufficient_quota`](/errors#402) until you add credit. See
[Going to production](/going-to-production#billing-readiness) for production
monitoring guidance.

For reserved capacity, custom limits, or invoicing beyond self-service
top-ups, [contact us](mailto:support@celeris.ai).