Congratulations on the launch! Personally would love to see a rough estimates of the expected number of requests and tokens required to run tasks like synthetic

Congratulations on the launch! Personally would love to see a rough estimates of the expected number of requests and tokens required to run tasks like synthetic data generation for different amounts of data. Though this is likely highly variable, would like to have a loose idea of possible incurred costs and execution time.