Understand and handle API rate limits

HTTP/1.1 200 OK
RateLimit-Limit: 100
RateLimit-Remaining: 95
RateLimit-Reset: 42          # seconds until the window resets
Retry-After: 42              # only on a 429

X-Monthly-Requests-Used: 1240
X-Requests-Limit: 8000
X-Requests-Remaining: 6760
X-Monthly-Tokens-Used: 402113
X-Tokens-Limit: 12000000

async function makeRequestWithRetry(url, options, maxRetries = 4) {
  for (let attempt = 0; attempt <= maxRetries; attempt++) {
    const response = await fetch(url, options);
    if (response.status !== 429 && response.status < 500) return response;
    if (attempt === maxRetries) return response; // give up — let the caller see it

    const retryAfter = Number(response.headers.get('Retry-After'));
    const waitMs = Number.isFinite(retryAfter) && retryAfter > 0
      ? retryAfter * 1000                                   // burst limit — server told us
      : Math.min(2 ** attempt * 1000, 30000) + Math.random() * 1000; // quota / 5xx — backoff + jitter

    console.log(`Retrying in ${Math.round(waitMs)}ms (attempt ${attempt + 1})`);
    await new Promise(r => setTimeout(r, waitMs));
  }
}

// Usage
const response = await makeRequestWithRetry(
  `${BASE_URL}/v1/chat/completions`,
  {
    method: 'POST',
    headers: {
      'Authorization': 'Bearer YOUR_API_KEY',
      'Content-Type': 'application/json'
    },
    body: JSON.stringify({
      model: 'gemini-2.5-flash',
      messages: [{ role: 'user', content: 'Hello!' }]
    })
  }
);