Rate Limits

​

How Rate Limiting Works

• You can have up to 10 requests processing at the same time
• Once a request completes, that slot becomes available immediately
• There’s no time window - it’s purely based on active concurrent requests

​

Example Scenario

Step 1: Send 10 requests
        → All 10 accepted ✓ (all slots occupied)

Step 2: Request #1 completes
        → 1 slot opens, can send 1 more

Step 3: Requests #2-4 complete
        → 3 slots open, can send 3 more

Step 4: Try to send 4 requests
        → 3 accepted, 1 rejected (at 10-slot limit)

​

Handling Rate Limits

{
  "error": {
    "code": "CONCURRENT_LIMIT_EXCEEDED",
    "message": "Concurrent Request Limit Reached",
    "details": "Maximum of 10 concurrent requests allowed. Please wait for existing requests to complete"
  }
}

​

Best Practices

• Keep track of how many requests are currently running
• Before sending a new request, check if you’re at the 10-request limit
• If at limit, wait for existing requests to complete before sending new ones
• Implement exponential backoff for 429 errors (wait 1s, 2s, 4s, 8s, etc.)

​

Need Higher Limits?