> ## 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, for each plan.

## OS3's models

All ids together. A minute's tokens count each request's whole prompt, cached part included, plus its output.

| Plan | A key, each minute | An organisation, each minute | Requests in flight | Fast id |
| - | - | - | - | - |
| Pay as you go | 300 requests, 24M tokens | 300 requests, 24M tokens | 8 | Yes |
| Pro | 600 requests, 60M tokens | 1,200 requests, 120M tokens | 32 | Yes |
| Enterprise | 1,200 requests, 120M tokens | 2,400 requests, 240M tokens | 32, or your contract's | Yes |

* Requests in flight are counted at each of OS3's gateway servers on its own, so an organisation whose requests reach several can have more in all.
* It is a cap, not reserved capacity: one organisation holds at most half of the normal lane on a GPU server, the same for every organisation.
* One agent that sends a full 1M-token prompt every 12 seconds uses about 5.3M tokens a minute.

## Your own provider's models

When OS3 sends your calls on to your own provider (OpenAI, Anthropic, Gemini and others), each key has its plan's limit, apart from the ones above:

| Plan | A key, each minute |
| - | - |
| Pay as you go | 300 requests, 2M tokens |
| Pro | 600 requests, 4M tokens |
| Enterprise | 1,200 requests, 8M tokens, or your contract's |

OS3 can set other limits on a key, or one limit over all your organisation's keys, on request.

## Past a limit

* Over requests or tokens a minute: `429 rate_limit_exceeded`, with `retry-after-ms` and `retry-after` headers. The OpenAI SDKs wait and retry by themselves.
* Over requests in flight: `429 tenant_busy`, with `Retry-After` (1 to 5 seconds).
* A stream that has started is never cut. Nothing refused is billed.

See [Errors](https://docs.os3compute.com/errors.md).


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