> For the complete documentation index, see [llms.txt](https://docs.kontinent.ai/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://docs.kontinent.ai/features/rate-limits.md).
# Rate limits & credits
## Rate limits
Each API key has an **RPM limit** (requests per minute) enforced over a sliding window of two 1-minute buckets. Exceeding it returns `429` with code `rate_limit_exceeded`. Back off and retry.
The limiter is **fail-open**: if the backing store has trouble, requests are never blocked. You will not be rate-limited due to an infrastructure hiccup on our side.
A separate `429 upstream_rate_limited` means the **provider** throttled us rather than your key hitting its limit. Retry with backoff.
| `429` code | Who hit a limit | What to do |
| ----------------------- | ----------------------- | ------------------------------------------ |
| `rate_limit_exceeded` | Your key's RPM | Back off; the sliding window is 60 s. |
| `upstream_rate_limited` | A provider throttled us | Retry with exponential backoff and jitter. |
## Credits
Kontinent is **prepaid**. Your organization holds a credit balance in [micro-EUR](/models-and-routing/models.md#pricing-and-the-money-unit), and each request debits the exact cost of the tokens it used. Metering runs asynchronously off the hot path, so billing never adds latency to your request.
:coins:
Pay per token
Billed on real usage, reconciled against an append-only ledger.
:receipt:
No hidden overhead
Metadata only: token counts, latency, status, model, and cost. Nothing more.
* Credits are checked **optimistically** against a hot balance, so a small overdraft (bounded by your RPM) is possible by design.
* A background worker reconciles spend within about **60 seconds**. Once an organization is exhausted, further requests get a hard `402` (`insufficient_credits`) until a top-up lands.
* A top-up **lifts suspension immediately**. You do not wait for the next reconcile cycle.
{% hint style="info" %}
`402 insufficient_credits` is safe to retry **after** topping up. Until then, retrying without adding credit will keep failing.
{% endhint %}
## Related
{% content-ref url="/pages/e4XLAToYQziKWGuYMiZu" %}
[Handling errors](/features/errors.md)
{% endcontent-ref %}
{% content-ref url="/pages/3TFSrUOiA2KRCa5V6cHt" %}
[Models & sovereignty](/models-and-routing/models.md)
{% endcontent-ref %}
---
# Agent Instructions
This documentation is published with GitBook. GitBook is the documentation platform designed so that both humans and AI agents can read, navigate, and reason over technical content effectively. Learn more at gitbook.com.
## Querying This Documentation
If you need additional information that is not directly available in this page, you can query the documentation dynamically by asking a question.
Perform an HTTP GET request on the current page URL with the `ask` query parameter, and the optional `goal` query parameter:
```
GET https://docs.kontinent.ai/features/rate-limits.md?ask=&goal=
```
`ask` is the immediate question: it should be specific, self-contained, and written in natural language.
`goal` is optional and describes the broader end goal you are ultimately trying to accomplish on behalf of the user. GitBook uses it to tailor the answer towards what is most useful for that goal.
The response will contain a direct answer to the question and relevant excerpts and sources from the documentation.
Use this mechanism when the answer is not explicitly present in the current page, you need clarification or additional context, or you want to retrieve related documentation sections.