Each service plan comes with a rate limit which sets the maximum number of messages that can be sent per second. The rate limit is calculated from all messages sent via the API. For example, a batch with 10 recipients will count as 10 messages for rate limiting purposes.
Each service plan gets its own message queue served in first-in-first-out order. This means that new batches will be accepted immediately but might be delayed if earlier batches are still in the queue.