List your company

Class-conditional Image Generation on ImageNet 256x256 2012 (train val)

1.26FID (w/o G)EDM2-L + DDO (downsampled*)
0.67924.59968.5212.4404Mar 3, 2025

Evaluation results

MethodLinks
EDM2-L + DDO (downsampled*)Type=Ours, #Params=777M, NFE=63, Generation Resolution=512x512, Resizing=OpenCV INTER_LANCZOS4 to 256x2562025.031.261.21
EDM2-L + DDO (downsampled*) + DPM-Solver-v3Type=Ours, #Params=777M, NFE=25, Generation Resolution=512x512, Resizing=OpenCV INTER_LANCZOS4 to 256x2562025.031.291.21
VAR-d30 (w/o sampling tricks) + DDOType=Ours, #Params=2.0B, NFE=10, sampling=no tricks2025.031.791.73
LightningDiTType=Diffusion, #Params=675M, NFE=2502025.032.171.35
VAR-d30Type=AR, #Params=2.0B, NFE=10, sampling=with top-k/top-p tricks2025.032.171.9
StyleGAN-XLType=GAN, #Params=166M, NFE=12025.032.3—
MAR-HType=Masked, #Params=943M, NFE=2562025.032.351.55
VDM++Type=Diffusion, #Params=2B, NFE=5122025.032.42.12
U-ViT, LType=Diffusion, #Params=2B, NFE=5122025.032.772.44
VAR-d16 (w/o sampling tricks) + DDOType=Ours, #Params=310M, NFE=10, sampling=no tricks2025.033.122.54
VAR-d30 (w/o sampling tricks) + CCAType=Ours, #Params=2.0B, NFE=10, sampling=no tricks2025.033.3—
RINType=Diffusion, #Params=410M, NFE=10002025.033.42—
VAR-d16Type=AR, #Params=310M, NFE=10, sampling=with top-k/top-p tricks2025.033.623.3
MAGVIT-v2Type=Masked, #Params=307M, NFE=642025.033.651.78
RAR-XXLType=AR, #Params=1.5B, NFE=2562025.033.911.48
ViT-VQGANType=AR, #Params=1.7B, NFE=10242025.034.17—
VAR-d30 (w/o sampling tricks)Type=Ours, #Params=2.0B, NFE=10, sampling=no tricks2025.034.741.92
SiT-XL + REPAType=Diffusion, #Params=675M, NFE=2502025.035.91.42
MaskGITType=Masked, #Params=227M, NFE=82025.036.18—
BigGAN-deepType=GAN, #Params=112M, NFE=12025.036.95—
ADM-UType=Diffusion, #Params=608M, NFE=5002025.037.493.94
RQ-TransformerType=AR, #Params=3.8B, NFE=682025.037.55—
SiT-XLType=Diffusion, #Params=675M, NFE=2502025.038.32.06
DiT-XL/2Type=Diffusion, #Params=675M, NFE=2502025.039.622.27
Open-MAGVIT2-XLType=AR, #Params=1.5B, NFE=2562025.039.632.33
LDM-4Type=Diffusion, #Params=400M, NFE=2502025.0310.563.6
ADMType=Diffusion, #Params=554M, NFE=2502025.0310.944.59
VAR-d16 (w/o sampling tricks)Type=Ours, #Params=310M, NFE=10, sampling=no tricks2025.0311.333.71
LlamaGen-3BType=AR, #Params=3.1B, NFE=2562025.0313.583.05
VQGANType=AR, #Params=1.4B, NFE=2562025.0315.78—
GigaGANType=GAN, #Params=569M, NFE=12025.03—3.45
MaskBitType=Masked, #Params=305M, NFE=2562025.03—1.52
xAR-HType=AR, #Params=1.1B, NFE=502025.03—1.24