List your company

Image Generation on ImageNet-1K 512x512 (val)

0.19rFIDUVIT-L/4
0.11880.59941.081.5606Jun 11, 2024Sep 27, 2024Jan 14, 2025May 3, 2025Aug 20, 2025Dec 7, 2025Mar 26, 2026

Evaluation results

MethodLinks
UVIT-L/4tokenizer=VAE [57], generator=UVIT-L/4 [4], parameters=287M, sampling_steps=502024.060.1918.0376.94.67213.31——
UVIT-H/4tokenizer=VAE [57], generator=UVIT-H/4 [4], parameters=501M, sampling_steps=502024.060.1915.71101.34.05263.80.6——
DiT-XL/2tokenizer=VAE [57], generator=DiT-XL/2 [49], parameters=675M, sampling_steps=2502024.060.1912.03105.33.04240.80.1——
UViT-L/4tokenizer=VAE ()‡, P=287M, S=50, T=1.02026.030.1918.0376.94.67213.3———
UViT-H/4tokenizer=VAE ()‡, P=501M, S=50, T=0.62026.030.1915.71101.34.05263.8———
DiT-XL/2tokenizer=VAE ()‡, P=675M, S=250, T=0.12026.030.1912.03105.33.04240.8———
CARD-Btokenizer=SMAP(KL)-B-128, P=234M, S=25, T=11.72026.030.694.29155.22.99253.1———
CARD-Ltokenizer=SMAP(KL)-B-128, P=568M, S=25, T=2.82026.030.693.12211.12.11298.4———
MAGVIT-v2tokenizer=LFQ [68], generator=MAGVIT-v2 [68], parameters=307M, sampling_steps=122024.061.224.61192.4——3.5——
MAGVIT-v2tokenizer=LFQ [68], generator=MAGVIT-v2 [68], parameters=307M, sampling_steps=642024.061.223.07213.11.91324.31——
MAGVIT-v2tokenizer=LFQ (Yu et al., 2024a), P=307M, S=12, T=3.52026.031.224.61192.4—————
MAGVIT-v2tokenizer=LFQ (Yu et al., 2024a), P=307M, S=64, T=1.02026.031.223.07213.11.91324.3———
TiTok-B-128tokenizer=TiTok-B-128, generator=MaskGIT-ViT [9], parameters=177M, sampling_steps=82024.061.373.911822.49260.433.3——
TiTok-B-128tokenizer=TiTok-B-128, generator=MaskGIT-ViT [9], parameters=177M, sampling_steps=642024.061.374.171812.13261.27.4——
MaskGIT-ViT (with SMAP-B-128)tokenizer=SMAP-B-128, P=177M, S=8, T=33.32026.031.373.911822.49260.4———
MaskGIT-ViT (with SMAP-B-128)tokenizer=SMAP-B-128, P=177M, S=64, T=7.42026.031.374.171812.13261.2———
TiTok-L-64tokenizer=TiTok-L-64, generator=MaskGIT-ViT [9], parameters=177M, sampling_steps=82024.061.783.64179.82.74221.141——
MaskGIT-ViT (with SMAP-L-64)tokenizer=SMAP-L-64, P=177M, S=8, T=41.02026.031.783.64179.82.74221.1———
MaskGIT-ViTtokenizer=MaskGIT-VQGAN [9], generator=MaskGIT-ViT [9], parameters=177M, sampling_steps=122024.061.977.32156——3.9——
MaskGIT-ViTtokenizer=MaskGIT-VQGAN (Chang et al., 2022), P=177M, S=12, T=3.92026.031.977.32156—————
BAR-L#params=1.1B, Guidance=classifier-free guidance (linear schedule)2026.02——————1.09311.1
DDT#params=675M, Guidance=classifier-free guidance2026.02——————1.28305.1
DiMR#params=525M, Guidance=classifier-free guidance2026.02——————2.89289.8
DiT#params=675M, Guidance=classifier-free guidance2026.02——————3.04240.8
MaskGiT#params=227M, Guidance=classifier-free guidance2026.02——————7.32156
RAE#params=839M, Guidance=classifier-free guidance2026.02——————1.13259.6
RAR#params=1.5B, Guidance=classifier-free guidance2026.02——————1.66295.7
REPA#params=675M, Guidance=classifier-free guidance2026.02——————2.08274.6
VAR#params=2.3B, Guidance=classifier-free guidance2026.02——————2.63303.2
VQGAN#params=227M, Guidance=classifier-free guidance2026.02——————26.5266.8
xAR#params=608M, Guidance=classifier-free guidance2026.02——————1.7281.5