One of the latest newcomers to the machine learning models developed by AI to generate digital images from natural language descriptions is MidJourney.
What is unique to it is that it uses the Discord UI to generate images and, of course, to monetise. Some said that the quality of images is lower compared to DALL-E or Imagen, but here is one my trials: “in a darkened room beyond the reach of gods faith, lies the wounded, the shattered remains of love betrayed*, — 8k render, hyperrealistic, rembrandt style”. Below is the result 👇