Rate limits and quotas

Two independent limits protect the platform: a per-second rate limit per API key (token bucket with 3x burst) and monthly quotas per application. Both return 429 rate_limit_error when exhausted.

Limits by plan

Plans also set webhook endpoint and team member counts — see pricing. Upgrades apply immediately; quotas reset on the first of each month (UTC).

How the rate limit works

• Each key has a token bucket holding 3x its rps in tokens; each request spends one.
• Tokens refill continuously at the plan's rps. Short spikes up to the burst size pass; sustained load above the rps drains the bucket and returns 429s.
• The limit is per key. Traffic split across keys of the same app is limited per key but metered against the app's shared monthly quota.

Rate limit headers

Handling 429s

Back off and retry — the bucket refills within seconds. A minimal decorrelated backoff:

async function withBackoff<T>(fn: () => Promise<T>, maxAttempts = 5): Promise<T> {
  let delay = 500;
  for (let attempt = 1; ; attempt++) {
    try {
      return await fn();
    } catch (err) {
      const retryable =
        err instanceof FeelConnectError &&
        (err.type === "rate_limit_error" || err.type === "api_error");
      if (!retryable || attempt === maxAttempts) throw err;
      await new Promise((r) => setTimeout(r, delay + Math.random() * delay));
      delay = Math.min(delay * 2, 8000);
    }
  }
}

const page = await withBackoff(() =>
  fc.transactions.sync({ access_token: token })
);

Monthly quotas

• API requests — every /v1 call counts one, including 4xx responses to authenticated requests.
• AI requests — AI product calls additionally count one AI request.
• At 80% of either quota, a usage.limit_approaching webhook fires and a banner appears in the dashboard.
• At 100%, further calls return 429 rate_limit_exceeded until the month resets or you upgrade. Data is never deleted — only the API gate closes.
• Track consumption in Dashboard → Usage, broken down per product and per day.