All are Worth Words: A ViT Backbone for Diffusion Models

Open paper

Benchmark	Model	Rank	Results
text-to-image-generation-on-coco	U-ViT-S/2-Deep	#2	FID: 5.48
text-to-image-generation-on-coco	U-ViT-S/2	#5	FID: 5.95