> ## Documentation Index
> Fetch the complete documentation index at: https://docs.poixe.com/llms.txt
> Use this file to discover all available pages before exploring further.

# Rate Limits

Poixe AI applies rate limits to API calls to keep the platform stable and ensure fair use. Rate limits are a **global, per-user quota**: all model requests under your account share the same rate allowance, which the system schedules centrally.

## What determines your rate limit

Rate limits are mainly tied to your **user level (Level)**. New users start at Level 0, and as your cumulative top-ups raise your level, you generally get higher request rates and more capacity.

You can view your level and the corresponding rate limits on the benefits page:
[https://poixe.com/tiers/user](https://poixe.com/tiers/user)

## Key concepts

* **RPM (Requests Per Minute)**: the maximum number of requests allowed per minute
* **Global quota**: requests across all models and all endpoints are counted together (there is no separate quota per model)
* **Takes effect immediately**: rate limits update automatically when your level increases, with no action required

## What happens when you hit a limit

If you send too many requests in a short period, the system may return rate-limit errors. We recommend that you:

* Add client-side retries (with exponential backoff)
* Queue or throttle concurrent requests
* For high-concurrency workloads, consider raising your level or contacting the team to evaluate your quota

## Check current status

In the dashboard you can view model status and capacity (useful when troubleshooting throttling or congestion):
[https://poixe.com/model](https://poixe.com/model)


This documentation is built and hosted on [Mintlify](https://mintlify.com), a developer documentation platform.