Skip to main content
Rate limits protect provider capacity, prevent accidental duplicate processing, and keep catalog imports predictable.

​
Current limits

20 images maximum per batch
25 images processed per hour per API key
1 active batch maximum per API key
2 active batches maximum per account
POST /api/v1/product-image-enhancements: 5 requests per hour per API key
GET /api/v1/product-image-enhancements/{enhancementId}: 120 requests per hour per API key
General API IP limit: 100 requests per hour
API key management also has dashboard-side limits: 
5 active API keys per account
5 key creations per hour
20 key creations per day
10 key mutations per hour

​
429 responses

Rate limit responses use HTTP 429. 
{
  "error": {
    "code": "rate_limit_exceeded",
    "message": "Rate limit exceeded.",
    "request_id": "req_..."
  }
}
Responses may include: 
Retry-After
X-RateLimit-Limit
X-RateLimit-Remaining
X-RateLimit-Reset
Wait for Retry-After before retrying.

​
Concurrency guidance

The API currently allows one active batch per API key and two active batches per account. If a platform needs to process a large catalog, split it into batches and enqueue the next batch only after the current one reaches a terminal status. Terminal enhancement statuses are: 
completed
partially_completed
failed
cancelled

​
Polling guidance

For low-volume integrations, poll every 5 to 15 seconds while an enhancement is queued or processing. For higher-volume workflows, use webhooks and keep polling as a fallback source of truth. Do not poll every item individually. Poll the enhancement status endpoint and read all item statuses from the response.