Text to Image AI Models: Different methodologies and different models, how it works?