List your company

Image Generation on ImageNet 512x512 (val)

1.366FID-50KSiD2A
1.286521.823012.35952.89599Feb 21, 2023Aug 11, 2023Jan 30, 2024Jul 19, 2024Jan 7, 2025Jun 27, 2025Dec 16, 2025

Evaluation results

MethodLinks
SiD2ACFG=N, NFE=1, Model Scale=XXL, Parameters=1.5B2024.101.366——————
SiD2ACFG=N, NFE=1, Model Scale=XL, Parameters=1.1B2024.101.379——————
EDM2-XXL (interval)NFE=632024.101.4——————
EDM2-XXLNFE=32×1.2, #Params=1.5B, Guidance Interval=true2025.031.4——————
SiD2ACFG=N, NFE=1, Model Scale=L, Parameters=777M2024.101.413——————
SiDACFG=N, NFE=1, Model Scale=XL, Parameters=1.1B2024.101.428——————
SiD2, Flop HeavyNFE=5122024.101.48——————
SiD2ACFG=N, NFE=1, Model Scale=M, Parameters=498M2024.101.488——————
SiD2, Flop Heavy distilledNFE=162024.101.5——————
SiDACFG=N, NFE=1, Model Scale=L, Parameters=777M2024.101.501——————
SiDACFG=N, NFE=1, Model Scale=XXL, Parameters=1.5B2024.101.503——————
SiDACFG=N, NFE=1, Model Scale=M, Parameters=498M2024.101.546——————
SiD2ACFG=N, NFE=1, Model Scale=S, Parameters=280M2024.101.669——————
EDM2-S (interval)NFE=632024.101.68——————
xAR-L#params=608M2025.021.7281.5—————
MARNFE=256×2, #Params=481M2025.031.73——————
SiDACFG=N, NFE=1, Model Scale=S, Parameters=280M2024.101.756——————
eMIGM-LNFE=64×1.25, #Params=478M2025.031.77——————
PaGoDANFE=12024.101.8——————
EDM2CFG=Y, NFE=126, Model Scale=XXL, Parameters=1.5B2024.101.81——————
EDM2-XXLNFE=32×2, #Params=1.5B2025.031.81——————
EDM2CFG=Y, NFE=126, Model Scale=XL, Parameters=1.1B2024.101.85——————
EDM2CFG=Y, NFE=126, Model Scale=L, Parameters=777M2024.101.88——————
sCDCFG=Y, NFE=2, Model Scale=XXL, Parameters=1.5B2024.101.88——————
sCD-XXLNFE=2, #Params=1.5B2025.031.88——————
SiDCFG=N, NFE=1, Model Scale=XL, Parameters=1.1B2024.101.888——————
SiDCFG=N, NFE=1, Model Scale=L, Parameters=777M2024.101.907——————
EDM2CFG=N, NFE=63, Model Scale=XXL, Parameters=1.5B2024.101.91——————
MAGVIT-v2NFE=642024.101.91——————
sCDCFG=Y, NFE=2, Model Scale=XL, Parameters=1.1B2024.101.93——————
EDM2CFG=N, NFE=63, Model Scale=XL, Parameters=1.1B2024.101.96——————
SiDCFG=N, NFE=1, Model Scale=XXL, Parameters=1.5B2024.101.969——————
EDM2CFG=Y, NFE=126, Model Scale=M, Parameters=498M2024.102.01——————
sCDCFG=Y, NFE=2, Model Scale=L, Parameters=777M2024.102.04——————
EDM2CFG=N, NFE=63, Model Scale=L, Parameters=777M2024.102.06——————
SiDCFG=N, NFE=1, Model Scale=M, Parameters=498M2024.102.06——————
REPA#params=675M, Backbone=DINOv22025.022.08274.6—————
SiD2ACFG=N, NFE=1, Model Scale=XS, Parameters=125M2024.102.156——————
SiD2, smallNFE=5122024.102.19——————
eMIGM-LNFE=16×1.2, #Params=478M2025.032.19——————
SiDACFG=N, NFE=1, Model Scale=XS, Parameters=125M2024.102.228——————
EDM2CFG=Y, NFE=126, Model Scale=S, Parameters=280M2024.102.23——————
SiT-XL/2 + MGBackbone=SiT-XL/2, Guidance=Model-guidance (MG)2025.022.24276.94.0386—60—
EDM2CFG=N, NFE=63, Model Scale=M, Parameters=498M2024.102.25——————
sCDCFG=Y, NFE=2, Model Scale=M, Parameters=498M2024.102.26——————
sCDCFG=Y, NFE=1, Model Scale=XXL, Parameters=1.5B2024.102.28——————
eMIGM-BNFE=64×1.25, #Params=244M2025.032.31——————
DiP-XL/32Model Type=Pixel Generative Models, Params=631M, CFG=With CFG2025.112.31291.684.4884—58—
sCDCFG=Y, NFE=1, Model Scale=XL, Parameters=1.1B2024.102.4——————
StyleGAN-XLNFE=12024.102.4——————
StyleGAN-XLModel=GAN, Label fraction (#labels/class)=100%2023.022.41267.75—————
StyleGAN-XLNFE=1×22025.032.41——————
FlowDCN-XLModel Type=Latent Generative Models, Params=618M+86M, CFG=With CFG2025.112.44252.84.5384—54—
sCDCFG=Y, NFE=2, Model Scale=S, Parameters=280M2024.102.5——————
MaskDiT-GModel Type=Latent Generative Models, Params=675M+86M, CFG=With CFG2025.112.5256.35.183—56—
Large-DiTNFE=250×2, #Params=3B2025.032.52——————
sCDCFG=Y, NFE=1, Model Scale=L, Parameters=777M2024.102.55——————
EDM2CFG=N, NFE=63, Model Scale=S, Parameters=280M2024.102.56——————
SiT-XL/2Backbone=SiT-XL/22025.022.62252.24.1884—57—
SiT-XLNFE=250×2, #Params=675M2025.032.62——————
SiT-XLModel Type=Latent Generative Models, Params=675M+86M, CFG=With CFG2025.112.62252.24.1884—57—
VAR-d36#params=2.3B2025.022.63303.2—————
VAR-d36-sNFE=10×2, #Params=2.3B2025.032.63——————
FreqWarm (USiT-H + DC-AE-f64c128)Diffusion Model=USiT-H, Autoencoder=DC-AE-f64c128, Warm-up=w/ FreqWarm, Params (B)=0.50, CFG=true, CFG Scale=1.52025.112.64276.28—————
VDM++NFE=2502024.102.65——————
VDM++NFE=512×2, #Params=2B2025.032.65——————
VDM++Model Type=Pixel Generative Models, Params=2.46B, CFG=With CFG2025.112.65278.1—————
DiffiTNFE=2502024.102.67——————
SiDCFG=N, NFE=1, Model Scale=S, Parameters=280M2024.102.707——————
OUSAC w/o cacheSteps=50, CFG-Steps=9, MACs (T)=30.94, Latency (s)=12.912025.122.71228.34.3883.43—57—
OUSACSteps=50, CFG-Steps=9, MACs (T)=24.93, Latency (s)=11.282025.122.72228.64.1383.62—55.3—
sCDCFG=Y, NFE=1, Model Scale=M, Parameters=498M2024.102.75——————
FreqWarm (UViT-H + DC-AE-f64c128)Diffusion Model=UViT-H, Autoencoder=DC-AE-f64c128, Warm-up=w/ FreqWarm, Params (B)=0.50, CFG=true, CFG Scale=1.52025.112.76246.89—————
DiT-XL/2 + MGBackbone=DiT-XL/2, Guidance=Model-guidance (MG)2025.022.78257.24.8683—58—
eMIGM-BNFE=16×1.2, #Params=244M2025.032.78——————
FreqWarm (DiT-XL + DC-AE-f32c128)Diffusion Model=DiT-XL, Autoencoder=DC-AE-f32c128, Warm-up=w/ FreqWarm, Params (B)=0.67, CFG=true, CFG Scale=1.52025.112.87240.05—————
DiMR-XL/3R#params=525M2025.022.89289.8—————
EDM2CFG=Y, NFE=126, Model Scale=XS, Parameters=125M2024.102.91——————
DiT-XL + Asym-VAE-f8Diffusion Model=DiT-XL, Autoencoder=Asym-VAE-f8, Warm-up=None, Params (B)=0.67, CFG=true, CFG Scale=1.52025.112.97——————
DDIMSteps=1000, CFG-Steps=1000, MACs (T)=1049.1, Latency (s)=416.792025.122.99210.64.3883.17—55.6—
SD, U-ViT-LNFE=5122024.103.02——————
SimDiffNFE=512×2, #Params=2B2025.033.02——————
SiDModel Type=Pixel Generative Models, Params=2.00B, CFG=With CFG2025.113.02248.7—————
DiT-XL/2-GModel=Diff., Label fraction (#labels/class)=100%2023.023.04240.82—————
DiT-XL/2Backbone=DiT-XL/22025.023.04240.85.0284—54—
DiT-XL/2#params=675M2025.023.04240.8—————
DiT-XL/2NFE=2502024.103.04——————
DiT-XL + SD-VAE-f8c4Diffusion Model=DiT-XL, Autoencoder=SD-VAE-f8c4, Warm-up=None, Params (B)=0.67, CFG=true, CFG Scale=1.52025.113.04——————
DiT-XL/2NFE=250×2, #Params=675M2025.033.04——————
DiT-XLModel Type=Latent Generative Models, Params=675M+86M, CFG=With CFG2025.113.04240.85.0284—54—
sCDCFG=Y, NFE=1, Model Scale=S, Parameters=280M2024.103.07——————
DiT-XL + DC-AE-f32c128Diffusion Model=DiT-XL, Autoencoder=DC-AE-f32c128, Warm-up=None, Params (B)=0.67, CFG=true, CFG Scale=1.52025.113.11227.69—————
USiT-H + DC-AE-f64c128Diffusion Model=USiT-H, Autoencoder=DC-AE-f64c128, Warm-up=None, Params (B)=0.50, CFG=true, CFG Scale=1.52025.113.13207.66—————
FreqWarm (USiT-2B + DC-AE-f64c128)Diffusion Model=USiT-2B, Autoencoder=DC-AE-f64c128, Warm-up=w/ FreqWarm, Params (B)=1.58, CFG=true, CFG Scale=1.52025.113.18311.79—————
FreqWarm (USiT-H + DC-AE-f32c128)Diffusion Model=USiT-H, Autoencoder=DC-AE-f32c128, Warm-up=w/ FreqWarm, Params (B)=0.50, CFG=true, CFG Scale=1.52025.113.2244.21—————
DDIMSteps=50, CFG-Steps=50, MACs (T)=52.45, Latency (s)=20.862025.123.2203.84.5383.27—56.4—
UViT-H + DC-AE-f64c128Diffusion Model=UViT-H, Autoencoder=DC-AE-f64c128, Warm-up=None, Params (B)=0.50, CFG=true, CFG Scale=1.52025.113.23219.3—————
eMIGM-SNFE=64×1.25, #Params=132M2025.033.29——————
OUSAC w/o cacheSteps=30, CFG-Steps=8, MACs (T)=19.93, Latency (s)=8.22025.123.33214.84.8582.99—55.3—
SiDCFG=N, NFE=1, Model Scale=XS, Parameters=125M2024.103.353——————