> ## Documentation Index
> Fetch the complete documentation index at: https://docs.os3compute.com/llms.txt
> Use this file to discover all available pages before exploring further.

# Limits by plan

> Requests and tokens a minute, and requests in flight.

Across all model ids. Tokens count the whole prompt, cached part included, plus output.

Requests / tokens a minute for each key and each organisation, and requests in flight:

| Plan | Per key | Per organisation | In flight |
| - | - | - | - |
| Pay as you go | 300 / 24M | 300 / 24M | 8 |
| Pro | 600 / 60M | 1,200 / 120M | 32 |
| Enterprise | 1,200 / 120M | 2,400 / 240M | 32 |

Enterprise limits can be your contract's instead.

Every plan includes the fast id.

Requests in flight count per OS3 gateway server. Past a limit you get a `429` with a retry header; a started stream is never cut. Ask [support](https://docs.os3compute.com/support.md) for other limits.


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