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

# Plans, quotas and limits

> What each plan includes, how quotas and rate limits apply, and what the API returns when you hit them.

## Plans

| Plan           | Decisions / month | Rate limit  | Run history | Live decisions | API keys  | Extras                                                             |
| -------------- | ----------------- | ----------- | ----------- | -------------- | --------- | ------------------------------------------------------------------ |
| **Free**       | 100               | 30 req/min  | 7 days      | 3              | 1         | LLM fallback, templates                                            |
| **Starter**    | 5,000             | 60 req/min  | 30 days     | unlimited      | unlimited | Versioning, test sets, human review                                |
| **Pro**        | 50,000            | 120 req/min | 30 days     | unlimited      | unlimited | Strict PII mode, quota and fallback alerts, priority support       |
| **Scale**      | 500,000           | 600 req/min | 30 days     | unlimited      | unlimited | Choice of fallback model                                           |
| **Enterprise** | custom            | 600 req/min | 30 days     | unlimited      | unlimited | Signed DPA, zero data retention at TypeSafe, SLA, annual invoicing |

Prices are listed on [branchpilot.ai/pricing](https://branchpilot.ai/pricing), excluding VAT. Plans are managed in the app under **Settings → Billing**.

<Note>
  **Early access.** A limited number of early accounts get the Pro plan free for three months, in exchange for candid feedback. It is applied automatically at sign-up while seats remain; promo codes offer the same thing on request.
</Note>

## Promo codes

A code entered in **Settings → Billing → Have a code?** offers a plan for a number of months (for example three months of Pro). It applies immediately; the account returns to Free at the end of the period unless a subscription was started in the meantime. A code cannot be applied to an account with a paid subscription, and each code works once per account.

## Monthly quota

One decision request = one unit, whatever the engine used (Jev or the LLM fallback) and whatever the outcome (`ok`, `uncertain`, `pending`). Playground runs count too; error responses do not.

The quota resets on the first day of each calendar month (UTC). Usage is visible on the dashboard, and the account owner is emailed at 80 % and 100 %.

When the quota is reached, `POST /decide/{slug}` answers **429** with `code: "quota_exceeded"`:

```json theme={null}
{
  "error": {
    "code": "quota_exceeded",
    "message": "Monthly quota of 100 decisions reached",
    "details": {
      "used": 100,
      "quota": 100,
      "retry_after": 432000,
      "upgrade_url": "https://app.branchpilot.ai/settings#billing"
    }
  }
}
```

The `Retry-After` header carries the number of seconds until the next period. There is no overage billing: upgrade the plan to resume immediately.

## Rate limit

Each API key has a per-minute request budget set by the plan (per account for the playground). Every response of `POST /decide/{slug}` carries:

| Header                  | Meaning                             |
| ----------------------- | ----------------------------------- |
| `X-RateLimit-Limit`     | requests allowed per minute         |
| `X-RateLimit-Remaining` | requests left in the current minute |
| `X-RateLimit-Reset`     | seconds until the window resets     |

Beyond the limit, the API answers **429** with `code: "rate_limited"` and a `Retry-After` header:

```json theme={null}
{
  "error": {
    "code": "rate_limited",
    "message": "Rate limit of 30 requests per minute reached",
    "details": { "limit": 30, "retry_after": 41 }
  }
}
```

Connectors should wait `Retry-After` seconds and retry; Make and n8n do this with a simple error handler or a *Retry* setting on the module.

## Plan limits in the app

* **Live decisions**: on Free, at most 3 decisions can be published at the same time; publishing a fourth is refused with an explicit message.
* **API keys**: on Free, one active key; revoke it to create another.
* **Strict PII mode**: available from Pro. Standard mode (declared fields plus detected emails, phones, IBANs, URLs and dates) is available on every plan.

## Retention

Runs are purged automatically after 7 days (Free) or 30 days (paid plans). Decisions, versions and test cases are kept for the life of the account. Deleting the account from **Settings** removes everything immediately; invoices remain in Stripe as the law requires.