To optimize resources, we use a batch system — all generation requests are grouped together and processed simultaneously.
Every video will complete unless an error occurs. If something fails, you'll see the reason displayed. Rate limits exist, but since this service is completely free, occasional delays may happen.