Diffusion models seem like they're poised to completely replace GANs. They obviously work super well, and you don't have this super finicky minimax training problem.
https://www.microsoft.com/en-us/research/publication/manifol...
"""
"Advantages over Traditional GANs" : Thus, we observe that our model exhibits _better training stability_ and mode coverage.
"Why is Sampling from Denoising Diffusion Models so Slow?" : After training, we generate novel instances by sampling from noise and iteratively denoising it _in a few steps_ using our denoising diffusion GAN generator.
"""