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

# Limits and usage

> Alpha request limits and current usage

Your tenant has request, concurrency, document, and monthly input-token limits. `GET /v1/usage` shows the limits assigned to you and your current usage.

The example uses `BREADBOWL_API_URL` and `BREADBOWL_API_KEY` from the [quickstart](/quickstart).

```bash theme={null}
curl --fail-with-body \
  "$BREADBOWL_API_URL/v1/usage" \
  -H "Authorization: Bearer $BREADBOWL_API_KEY"
```

Unless onboarding assigned different values, the alpha defaults are:

| Limit                      |    Default |
| -------------------------- | ---------: |
| Requests per minute        |         60 |
| Concurrent requests        |          2 |
| Input tokens per UTC month | 50,000,000 |
| Stored documents           |    100,000 |

Request size limits are:

| Input                                 |   Maximum |
| ------------------------------------- | --------: |
| JSON body                             |     5 MiB |
| Documents per upload or score request |        50 |
| Text per document                     |   100 KiB |
| Query text                            |     8 KiB |
| `top_k` / `candidate_k`               | 100 / 500 |

Token values in the usage response are decimal strings. Use `BigInt` or another arbitrary-precision integer type if you calculate with them.

On HTTP 429, inspect `error.code` and honor `Retry-After`. Reduce concurrency for `concurrency_limit_exceeded`; stop retrying when monthly usage is exhausted.