HyperAIHyperAI

Command Palette

Search for a command to run...

Console

Image Generation On Imagenet 256X256

Metrics

FID

Results

Performance results of various models on this benchmark

Paper Title
Improved DDPM12.3Improved Denoising Diffusion Probabilistic Models
ADM11.84-
BigGAN-deep8.1Large Scale GAN Training for High Fidelity Natural Image Synthesis
Polarity-BigGAN6.82Polarity Sampling: Quality and Diversity Control of Pre-Trained Generative Networks via Singular Values
VQGAN+Transformer (k=mixed, p=1.0, a=0.005)6.59Taming Transformers for High-Resolution Image Synthesis
MaskGIT6.18MaskGIT: Masked Generative Image Transformer
VQGAN+Transformer (k=600, p=1.0, a=0.05)5.2Taming Transformers for High-Resolution Image Synthesis
CDM4.88Cascaded Diffusion Models for High Fidelity Image Generation
ADM-G4.59Diffusion Models Beat GANs on Image Synthesis
RIN4.51Scalable Adaptive Computation for Iterative Generation
LFM4.46Flow Matching in Latent Space
ADM-G++ (Recall)4.45Refining Generative Process with Discriminator Guidance in Score-based Diffusion Models
LDM4.29-
ADM-G + EDS + ECT (ED-DPM, classifier_scale=1.0)4.09Entropy-driven Sampling and Training Scheme for Conditional Diffusion Generation
MaskGIT (a=0.05)4.02MaskGIT: Masked Generative Image Transformer
ADM-G + EDS (ED-DPM, classifier_scale=0.75)3.96Entropy-driven Sampling and Training Scheme for Conditional Diffusion Generation
ADM-G, ADM-U3.94Diffusion Models Beat GANs on Image Synthesis
RQ-Transformer3.83Autoregressive Image Generation using Residual Quantization
simple diffusion (U-ViT, L)3.75Simple diffusion: End-to-end diffusion for high resolution images
simple diffusion (U-Net)3.71Simple diffusion: End-to-end diffusion for high resolution images
0 of 79 row(s) selected.
Image Generation On Imagenet 256X256 | SOTA | HyperAI