List your company

Conditional Image Generation on ImageNet 512x512 (val)

1.52gFIDMacTok+SiT-XL‡
0.527.2714.0220.77Dec 14, 2024Mar 2, 2025May 20, 2025Aug 7, 2025Oct 24, 2025Jan 11, 2026Mar 31, 2026

Evaluation results

MethodLinks
MacTok+SiT-XL‡# Params (G)=675M, Tok. Model=KL, Token Type=1D, # Params (T)=391M†, # Tokens=64, Classifier-Free Guidance (CFG)=true2026.031.52306——0.89——
MacTok+SiT-XL‡# Params (G)=675M, Tok. Model=KL, Token Type=1D, # Params (T)=176M, # Tokens=128, Classifier-Free Guidance (CFG)=true2026.031.52316——0.79——
MacTok+SiT-XL# Params(G)=675M, Tok. Model=KL, Token Type=1D, # Params(T)=391M, # Tokens=64, Tok. rFID=0.89, CFG setting=w/ CFG2026.031.523060.80.63———
MacTok+SiT-XL# Params(G)=675M, Tok. Model=KL, Token Type=1D, # Params(T)=176M, # Tokens=128, Tok. rFID=0.79, CFG setting=w/ CFG2026.031.523160.80.63———
MAETok‡# Params (G)=675M, Tok. Model=AE, Token Type=1D, # Params (T)=176M, # Tokens=128, Classifier-Free Guidance (CFG)=true2026.031.69304.2——0.62——
MAETok# Params(G)=675M, Tok. Model=AE, Token Type=1D, # Params(T)=176M, # Tokens=128, Tok. rFID=0.62, CFG setting=w/ CFG2026.031.69304.20.820.62———
MAR-LType=Mask., Estimated inference time=via linear scaling2026.021.73279.9———214.4—
MAR-H# Params (G)=943M, Tok. Model=KL, Token Type=2D, # Params (T)=66M, # Tokens=1024, Classifier-Free Guidance (CFG)=true2026.031.73279.9—————
MAR-H# Params(G)=943M, Tok. Model=KL, Token Type=2D, # Params(T)=66M, # Tokens=1024, CFG setting=w/ CFG2026.031.73279.90.770.61———
TexTok-128# Params (G)=675M, Tok. Model=KL, Token Type=1D, # Params (T)=176M, # Tokens=128, Classifier-Free Guidance (CFG)=true2026.031.8305.4——0.97——
TexTok-128# Params(G)=675M, Tok. Model=KL, Token Type=1D, # Params(T)=176M, # Tokens=128, Tok. rFID=0.97, CFG setting=w/ CFG2026.031.8305.40.810.63———
MAGVIT-v2# Params (G)=307M, Tok. Model=LFQ, Token Type=2D, # Params (T)=116M, # Tokens=1024, Classifier-Free Guidance (CFG)=true2026.031.91324.3—————
MAGVIT-v2# Params(G)=307M, Tok. Model=LFQ, Token Type=2D, # Params(T)=116M, # Tokens=1024, CFG setting=w/ CFG2026.031.91324.3—————
TiTok-BTok. Model=VQ, # Tokens=128, CFG=true2024.122.13261.2——1.52——
TiTok-B-128# Params (G)=177M, Tok. Model=VQ, Token Type=1D, # Params (T)=202M, # Tokens=128, Classifier-Free Guidance (CFG)=true2026.032.13261.2——1.52——
TiTok-B-128# Params(G)=177M, Tok. Model=VQ, Token Type=1D, # Params(T)=202M, # Tokens=128, Tok. rFID=1.52, CFG setting=w/ CFG2026.032.13261.2—————
SiT-XLTok. Model=SoftVQ-BL, # Tokens=64, Gflops=29.23, Throughput (imgs/sec)=5.38, CFG=true2024.122.21290.50.850.590.71——
SoftVQ-VAE‡# Params (G)=675M, Tok. Model=SoftVQ, Token Type=1D, # Params (T)=391M, # Tokens=64, Classifier-Free Guidance (CFG)=true2026.032.21290.5——0.71——
SoftVQ-VAE# Params(G)=675M, Tok. Model=SoftVQ, Token Type=1D, # Params(T)=391M, # Tokens=64, Tok. rFID=0.71, CFG setting=w/ CFG2026.032.21290.50.850.59———
UViT-2BGflops=104.18, Throughput (imgs/sec)=9.44, Sampler=DPM-Solver, Sampling steps=50, CFG=true2024.122.25——————
UViT-2B# Params (G)=2B, Tok. Model=AE, Token Type=2D, # Params (T)=323M, # Tokens=256, Classifier-Free Guidance (CFG)=true2026.032.25———0.22——
UViT-2B# Params(G)=2B, Tok. Model=AE, Token Type=2D, # Params(T)=323M, # Tokens=256, Tok. rFID=0.22, CFG setting=w/ CFG2026.032.25——————
VAR-d36 + SSGType=VAR2026.022.39320.6———1—
StyleGAN-XLCFG=true2024.122.41267.7—————
StyleGAN-XL# Params (G)=168M, Classifier-Free Guidance (CFG)=true2026.032.41267.7—————
StyleGAN-XL# Params(G)=168M, CFG setting=w/ CFG2026.032.41267.7—————
UViT-HGflops=128.9, Throughput (imgs/sec)=6.14, Sampler=DPM-Solver, Sampling steps=50, CFG=true2024.122.53——————
UViT-H# Params (G)=501M, Tok. Model=KL, Token Type=2D, # Params (T)=84M, # Tokens=4096, Classifier-Free Guidance (CFG)=true2026.032.53———0.62——
UViT-H# Params(G)=501M, Tok. Model=KL, Token Type=2D, # Params(T)=84M, # Tokens=4096, Tok. rFID=0.62, CFG setting=w/ CFG2026.032.53——————
SiT-XL/2Gflops=373.32, Throughput (imgs/sec)=0.1, CFG=true2024.122.62252.20.840.570.1——
SiT-XL/2# Params (G)=675M, Token Type=2D, Classifier-Free Guidance (CFG)=true2026.032.62252.2—————
SiT-XL/2# Params(G)=675M, Token Type=2D, CFG setting=w/ CFG2026.032.62252.20.840.57———
UViT-HTok. Model=AE, # Tokens=64, Gflops=32.99, Throughput (imgs/sec)=15.23, Sampler=DPM-Solver, Sampling steps=50, CFG=true2024.122.66———0.22——
UViT-H# Params (G)=501M, Token Type=2D, Classifier-Free Guidance (CFG)=true2026.032.66——————
UViT-H# Params(G)=501M, Token Type=2D, CFG setting=w/ CFG2026.032.66——————
DiffiTType=Diff.2026.022.67252.1—————
VAR-d36Type=VAR, Status=Reproduced2026.022.7290.6———1—
MAR-H# Params (G)=943M, Tok. Model=KL, Token Type=2D, # Params (T)=66M, # Tokens=1024, Classifier-Free Guidance (CFG)=false2026.032.74205.2—————
TiTok-L-64# Params (G)=177M, Tok. Model=VQ, Token Type=1D, # Params (T)=644M, # Tokens=64, Classifier-Free Guidance (CFG)=true2026.032.74221.1——1.77——
MAR-H# Params(G)=943M, Tok. Model=KL, Token Type=2D, # Params(T)=66M, # Tokens=1024, CFG setting=w/o CFG2026.032.74205.20.690.59———
TiTok-L-64# Params(G)=177M, Tok. Model=VQ, Token Type=1D, # Params(T)=644M, # Tokens=64, Tok. rFID=1.77, CFG setting=w/ CFG2026.032.74221.1—————
MAETok‡# Params (G)=675M, Tok. Model=AE, Token Type=1D, # Params (T)=176M, # Tokens=128, Classifier-Free Guidance (CFG)=false2026.032.79204.3——0.62——
MAETok# Params(G)=675M, Tok. Model=AE, Token Type=1D, # Params(T)=176M, # Tokens=128, Tok. rFID=0.62, CFG setting=w/o CFG2026.032.79204.30.810.62———
DiT-XLTok. Model=AE, # Tokens=256, Gflops=80.75, Throughput (imgs/sec)=1.02, CFG=true2024.122.84———0.22——
DiT-XL# Params (G)=675M, Token Type=2D, Classifier-Free Guidance (CFG)=true2026.032.84——————
DiT-XL# Params(G)=675M, Token Type=2D, CFG setting=w/ CFG2026.032.84——————
HMAR-d24Type=VAR2026.022.99304.1—————
DiT-XL/2Gflops=373.34, Throughput (imgs/sec)=0.1, CFG=true2024.123.04240.80.840.540.1——
DiT-XL/2Type=Diff.2026.023.04240.8———81—
DiT-XL/2# Params (G)=675M, Token Type=2D, Classifier-Free Guidance (CFG)=true2026.033.04240.8—————
DiT-XL/2# Params(G)=675M, Token Type=2D, CFG setting=w/ CFG2026.033.04240.80.840.54———
DIFFUSSM-XL-GType=Diff.2026.023.41255.1—————
MAR-HTok. Model=SoftVQ-BL, # Tokens=64, Gflops=86.55, Throughput (imgs/sec)=1.5, CFG=true2024.123.42261.80.770.610.71——
ADMCFG=true2024.123.85221.70.840.53———
ADMClassifier-Free Guidance (CFG)=true2026.033.85221.7—————
ADMCFG setting=w/ CFG2026.033.85221.70.840.53———
U-ViT-H/4Tok. Model=KL, # Tokens=4096, Gflops=128.92, Throughput (imgs/sec)=0.58, Sampler=DPM-Solver, Sampling steps=50, CFG=true2024.124.05263.80.840.480.62——
U-ViT-H/4# Params (G)=501M, Token Type=2D, Classifier-Free Guidance (CFG)=true2026.034.05263.8—————
U-ViT-H/4# Params(G)=501M, Token Type=2D, CFG setting=w/o CFG2026.034.05263.80.840.48———
SiT-XLTok. Model=SoftVQ-L, # Tokens=32, Gflops=14.73, Throughput (imgs/sec)=10.12, CFG=true2024.124.232180.830.520.64——
MacTok+SiT-XL‡# Params (G)=675M, Tok. Model=KL, Token Type=1D, # Params (T)=391M†, # Tokens=64, Classifier-Free Guidance (CFG)=false2026.034.63163.7——0.89——
MacTok+SiT-XL# Params(G)=675M, Tok. Model=KL, Token Type=1D, # Params(T)=391M, # Tokens=64, Tok. rFID=0.89, CFG setting=w/o CFG2026.034.63163.70.80.61———
MacTok+SiT-XL‡# Params (G)=675M, Tok. Model=KL, Token Type=1D, # Params (T)=176M, # Tokens=128, Classifier-Free Guidance (CFG)=false2026.035.12156.3——0.79——
MacTok+SiT-XL# Params(G)=675M, Tok. Model=KL, Token Type=1D, # Params(T)=176M, # Tokens=128, Tok. rFID=0.79, CFG setting=w/o CFG2026.035.12156.30.790.61———
UViT-2B# Params (G)=2B, Tok. Model=AE, Token Type=2D, # Params (T)=323M, # Tokens=256, Classifier-Free Guidance (CFG)=false2026.036.5———0.22——
UViT-2B# Params(G)=2B, Tok. Model=AE, Token Type=2D, # Params(T)=323M, # Tokens=256, Tok. rFID=0.22, CFG setting=w/o CFG2026.036.5——————
MaskGITTok. Model=VQ, # Tokens=1024, CFG=false2024.127.32156——1.97——
MaskGITType=Mask.2026.027.32156———0.5—
MaskGIT# Params (G)=227M, Tok. Model=VQ, Token Type=2D, # Params (T)=66M, # Tokens=1024, Classifier-Free Guidance (CFG)=false2026.037.32156——1.97——
MaskGIT# Params(G)=227M, Tok. Model=VQ, Token Type=2D, # Params(T)=66M, # Tokens=1024, Tok. rFID=1.97, CFG setting=w/o CFG2026.037.32156—————
SiT-XLTok. Model=SoftVQ-BL, # Tokens=64, Gflops=29.23, Throughput (imgs/sec)=5.38, CFG=false2024.127.96133.90.730.630.71——
SoftVQ-VAE‡# Params (G)=675M, Tok. Model=SoftVQ, Token Type=1D, # Params (T)=391M, # Tokens=64, Classifier-Free Guidance (CFG)=false2026.037.96133.9——0.71——
SoftVQ-VAE# Params(G)=675M, Tok. Model=SoftVQ, Token Type=1D, # Params(T)=391M, # Tokens=64, Tok. rFID=0.71, CFG setting=w/o CFG2026.037.96133.90.730.63———
MAR-HTok. Model=SoftVQ-BL, # Tokens=64, Gflops=86.55, Throughput (imgs/sec)=1.5, CFG=false2024.128.21152.90.690.590.71——
BigGANCFG=true2024.128.43177.9—————
BigGANType=GAN2026.028.43177.9—————
BigGANClassifier-Free Guidance (CFG)=true2026.038.43177.9—————
BigGANCFG setting=w/ CFG2026.038.43177.9—————
DiT-XL# Params (G)=675M, Token Type=2D, Classifier-Free Guidance (CFG)=false2026.039.56——————
DiT-XL# Params(G)=675M, Token Type=2D, CFG setting=w/o CFG2026.039.56——————
DiT-XL/2Gflops=373.34, Throughput (imgs/sec)=0.1, CFG=false2024.129.62121.5——0.1——
DiT-XL/2# Params (G)=675M, Token Type=2D, Classifier-Free Guidance (CFG)=false2026.039.62121.5—————
DiT-XL/2# Params(G)=675M, Token Type=2D, CFG setting=w/o CFG2026.039.62121.5—————
UViT-H# Params (G)=501M, Tok. Model=KL, Token Type=2D, # Params (T)=84M, # Tokens=4096, Classifier-Free Guidance (CFG)=false2026.039.83———0.62——
UViT-H# Params(G)=501M, Tok. Model=KL, Token Type=2D, # Params(T)=84M, # Tokens=4096, Tok. rFID=0.62, CFG setting=w/o CFG2026.039.83——————
SiT-XLTok. Model=SoftVQ-L, # Tokens=32, Gflops=14.73, Throughput (imgs/sec)=10.12, CFG=false2024.1210.17119.20.650.590.64——
UViT-H# Params (G)=501M, Token Type=2D, Classifier-Free Guidance (CFG)=false2026.0312.26——————
UViT-H# Params(G)=501M, Token Type=2D, CFG setting=w/o CFG2026.0312.26——————
ADMCFG=false2024.1223.2458.06—————
ADMClassifier-Free Guidance (CFG)=false2026.0323.2458.1—————
ADMCFG setting=w/o CFG2026.0323.2458.1—————
VQGANType=AR2026.0226.5266.8———25—
EDM2-LNFE=632026.03—250.070.7530.61——67.7
EDM2-L + CC-DPONFE=632026.03—298.890.8120.544——51.92
EDM2-L + CCANFE=632026.03—281.450.790.574——51.45
EDM2-L + CFGNFE=1262026.03—360.30.8440.512——39.86
EDM2-L + DDONFE=632026.03—268.720.7370.652——49.47
EDM2-L + MCLRNFE=632026.03—332.020.8490.492——42.5