Explain the problem as you see it
As new models get added, it would be helpful to be able to generate images with some of the more powerful models (imagen 3 or 4, ideogram).
Why is this a problem for you?
For me it's all about reducing the context switching...I go to midjourney to make images and then have to bring them in...inline generation would be spectacular
Suggest a solution
You Are already connection to many models, it doesn't seem like adding an image based one would be that difficult.