Vision Foundation Models as Generalist Tokenizers for Image Generation

Benchmark Model Rank Results
image-generation-on-imagenet-256x256VFMTok-2B(SigLIP)#101FID: 2.05
image-generation-on-imagenet-256x256VFMTok-2B(DINOv2)#106FID: 2.09
image-generation-on-imagenet-256x256VFMTok-2B(SigLIP2)#116FID: 2.17
image-generation-on-imagenet-256x256VFMTok-XL(DINOv2)#131FID: 2.41