I am super hopeful! Hardware is improving, inference costs will continue to decrease, models will only improve...