MAGVIT-v2 (w/o guidance)

Reported on 3 benchmarks across 1 task · 1 paper

Note: results are matched by exact model name. Different papers may use the same name for different model variants.

Medical3 results

Image GenerationonImageNet 512x512
FID· 2023-10-09
3.07
best: 1.21 (EDM2-L + DDO (SD-VAE, 25 steps, DPM-Solver-v3))
Language Model Beats Diffusion -- Tokenizer is Key to Visual Generation arXiv:2310.05737
Image GenerationonImageNet 512x512
Inception score· 2023-10-09
213.1
best: 350.2 (DPC-U)
Language Model Beats Diffusion -- Tokenizer is Key to Visual Generation arXiv:2310.05737
Image GenerationonImageNet 256x256
FID· 2023-10-09
3.65
best: 1.06 (SiT-XL/2 + UCGM-S (E2E-VAE + 40 sampling steps + CFG))
Language Model Beats Diffusion -- Tokenizer is Key to Visual Generation arXiv:2310.05737