Rate Limits

To ensure high availability and fair usage across the platform, ChatboxAI enforces rate limits at the API key and Account level.

1. Free Tier Limits

The Free tier is designed for hobbyists, students, and developers testing the platform before deploying to production.

• Global Daily Limit: Free accounts are strictly limited to 300 requests per day across all endpoints.
• Reset Time: The daily limit resets at midnight UTC.
• Concurrency: Free accounts are limited to 2 concurrent requests.
• Model Restrictions: Only Free-tier models (e.g., chatboxai/llama-3.3-70b, chatboxai/qwen3-32b) are accessible.

If you exceed the 300 requests/day limit, the API will return a 429 Too Many Requests error with a JSON body indicating that the daily quota has been exhausted.

2. Paid Tier Limits

Paid tier limits are significantly higher and are designed for production workloads. They operate on a Requests Per Minute (RPM) and Tokens Per Minute (TPM) basis rather than a strict daily hardcap.

Note: Paid limits can be increased by contacting support once your account has established a consistent billing history.

3. Handling Rate Limit Errors (HTTP 429)

When a rate limit is exceeded, ChatboxAI returns a 429 status code. You should implement an exponential backoff strategy in your application to handle these gracefully.

Response Headers

When you receive a 429, or even on successful 200 responses, check the following headers to monitor your quota:

• X-RateLimit-Limit-Requests: The maximum RPM allowed.
• X-RateLimit-Remaining-Requests: How many requests you have left in the current minute window.
• X-RateLimit-Reset-Requests: Time in seconds until the RPM counter resets.

Example Retry Logic (Node.js)

async function fetchWithRetry(url, options, retries = 3, backoff = 1000) {
  for (let i = 0; i < retries; i++) {
    const res = await fetch(url, options);
    
    if (res.status === 429) {
      console.warn(`Rate limited. Retrying in ${backoff}ms...`);
      await new Promise(r => setTimeout(r, backoff));
      backoff *= 2; // Exponential backoff (1s -> 2s -> 4s)
      continue;
    }
    
    return res;
  }
  throw new Error("Max retries exceeded");
}

4. Insufficient Credits (HTTP 402)

If you are on the Paid tier and your prepaid credit balance drops to $0.00, the API will stop serving requests and return a 402 Payment Required error.

To prevent production outages, we highly recommend enabling Auto-Recharge in the Billing section of your dashboard.