No

The comment was directed at “doesn't this method add another cost and overhead for calling Text-to-Image models”