Image Generation on ImageNet 64x64

Metric: FID (lower is better)

LeaderboardDataset

Loading chart...

Results

Sort:

#	Model↕	FID▲	Extra Data	Paper	Date↕	Code
1	SIMS	0.92	No	Self-Improving Diffusion Models with Synthetic D...	2024-08-29	-
2	EDM2-S+DDO	0.97	No	Direct Discriminative Optimization: Your Likelih...	2025-03-03	Code
3	Uni-Instruct	1.02	No	-	-	-
4	SiDA-EDM	1.11	No	Adversarial Score identity Distillation: Rapidly...	2024-10-19	Code
5	GDD-I	1.16	No	Diffusion Models Are Innate One-Step Generators	2024-05-31	Code
6	PaGoDA	1.21	No	PaGoDA: Progressive Growing of a One-Step Genera...	2024-05-23	Code
7	DisCo-Diff	1.22	No	DisCo-Diff: Enhancing Continuous Diffusion Model...	2024-07-03	Code
8	RIN	1.23	No	Scalable Adaptive Computation for Iterative Gene...	2022-12-22	Code
9	GDD	1.42	No	Diffusion Models Are Innate One-Step Generators	2024-05-31	Code
10	SCT	1.47	No	Stable Consistency Tuning: Understanding and Imp...	2024-10-24	Code
11	CDM	1.48	No	Cascaded Diffusion Models for High Fidelity Imag...	2021-05-30	-
12	StyleGAN-XL	1.51	No	StyleGAN-XL: Scaling StyleGAN to Large Diverse D...	2022-02-01	Code
13	SiD	1.524	No	Score identity Distillation: Exponentially Fast ...	2024-04-05	Code
14	TCM	1.62	No	Truncated Consistency Models	2024-10-18	-
15	ECM-XL	1.67	No	Consistency Models Made Easy	2024-06-20	Code
16	CAF	1.69	No	Constant Acceleration Flow	2024-11-01	Code
17	CTM	1.73	No	Consistency Trajectory Models: Learning Probabil...	2023-10-01	Code
18	ADM (dropout)	2.07	No	Diffusion Models Beat GANs on Image Synthesis	2021-05-11	Code
19	LEGO	2.16	No	Learning Stackable and Skippable LEGO Bricks for...	2023-10-10	Code
20	TarFlow	2.9	No	Normalizing Flows are Capable Generative Models	2024-12-09	Code
21	Improved DDPM	2.92	No	Improved Denoising Diffusion Probabilistic Models	2021-02-18	Code
22	2-rectified flow++ (NFE=2)	3.64	No	Improving the Training of Rectified Flows	2024-05-30	Code
23	2-rectified flow++ (NFE=1)	4.31	No	Improving the Training of Rectified Flows	2024-05-30	Code
24	CD (Diffusion + Distillation, NFE=2)	4.7	No	Consistency Models	2023-03-02	Code
25	CD (Diffusion + Distillation, NFE=1)	6.2	No	Consistency Models	2023-03-02	Code
26	IC-GAN + DA	6.7	No	Instance-Conditioned GAN	2021-09-10	Code
27	SAGAN + instance selection	9.07	No	Instance Selection for GANs	2020-07-30	Code
28	FQ-GAN	9.67	No	Feature Quantization Improves GAN Training	2020-04-05	Code
29	CT (Direct Generation, NFE=2)	11.1	No	Consistency Models	2023-03-02	Code
30	CT (Direct Generation, NFE=1)	13	No	Consistency Models	2023-03-02	Code
31	FM	14.45	No	Flow Matching for Generative Modeling	2022-10-06	Code
32	CLR-GAN	20.27	No	-	-	Code
33	PGMGAN	21.73	No	Partition-Guided GANs	2021-04-02	Code
34	GLIDE + CLIP + CLS + CLS-FREE	29.184	No	Composing Ensembles of Pre-trained Models via It...	2022-10-20	-
35	GLIDE + CLS-FREE	29.219	No	Composing Ensembles of Pre-trained Models via It...	2022-10-20	-
36	GLIDE + CLIP	30.462	No	Composing Ensembles of Pre-trained Models via It...	2022-10-20	-
37	GLIDE + CLS	30.871	No	Composing Ensembles of Pre-trained Models via It...	2022-10-20	-

#1SIMSSOTA
0.92
FID· 2024-08-29
Self-Improving Diffusion Models with Synthetic Data
#2EDM2-S+DDO
0.97
FID· 2025-03-03
Direct Discriminative Optimization: Your Likelihood-Based Visual Generative Model is Secretly a GAN Discriminator Code
#3Uni-Instruct
1.02
FID
No paper
#4SiDA-EDM
1.11
FID· 2024-10-19
Adversarial Score identity Distillation: Rapidly Surpassing the Teacher in One Step Code
#5GDD-ISOTA
1.16
FID· 2024-05-31
Diffusion Models Are Innate One-Step Generators Code
#6PaGoDASOTA
1.21
FID· 2024-05-23
PaGoDA: Progressive Growing of a One-Step Generator from a Low-Resolution Diffusion Teacher Code
#7DisCo-Diff
1.22
FID· 2024-07-03
DisCo-Diff: Enhancing Continuous Diffusion Models with Discrete Latents Code
#8RINSOTA
1.23
FID· 2022-12-22
Scalable Adaptive Computation for Iterative Generation Code
#9GDD
1.42
FID· 2024-05-31
Diffusion Models Are Innate One-Step Generators Code
#10SCT
1.47
FID· 2024-10-24
Stable Consistency Tuning: Understanding and Improving Consistency Models Code
#11CDMSOTA
1.48
FID· 2021-05-30
Cascaded Diffusion Models for High Fidelity Image Generation
#12StyleGAN-XL
1.51
FID· 2022-02-01
StyleGAN-XL: Scaling StyleGAN to Large Diverse Datasets Code
#13SiD
1.524
FID· 2024-04-05
Score identity Distillation: Exponentially Fast Distillation of Pretrained Diffusion Models for One-Step Generation Code
#14TCM
1.62
FID· 2024-10-18
Truncated Consistency Models
#15ECM-XL
1.67
FID· 2024-06-20
Consistency Models Made Easy Code
#16CAF
1.69
FID· 2024-11-01
Constant Acceleration Flow Code
#17CTM
1.73
FID· 2023-10-01
Consistency Trajectory Models: Learning Probability Flow ODE Trajectory of Diffusion Code
#18ADM (dropout)SOTA
2.07
FID· 2021-05-11
Diffusion Models Beat GANs on Image Synthesis Code
#19LEGO
2.16
FID· 2023-10-10
Learning Stackable and Skippable LEGO Bricks for Efficient, Reconfigurable, and Variable-Resolution Diffusion Modeling Code
#20TarFlow
2.9
FID· 2024-12-09
Normalizing Flows are Capable Generative Models Code
#21Improved DDPMSOTA
2.92
FID· 2021-02-18
Improved Denoising Diffusion Probabilistic Models Code
#222-rectified flow++ (NFE=2)
3.64
FID· 2024-05-30
Improving the Training of Rectified Flows Code
#232-rectified flow++ (NFE=1)
4.31
FID· 2024-05-30
Improving the Training of Rectified Flows Code
#24CD (Diffusion + Distillation, NFE=2)
4.7
FID· 2023-03-02
Consistency Models Code
#25CD (Diffusion + Distillation, NFE=1)
6.2
FID· 2023-03-02
Consistency Models Code
#26IC-GAN + DA
6.7
FID· 2021-09-10
Instance-Conditioned GAN Code
#27SAGAN + instance selectionSOTA
9.07
FID· 2020-07-30
Instance Selection for GANs Code
#28FQ-GANSOTA
9.67
FID· 2020-04-05
Feature Quantization Improves GAN Training Code
#29CT (Direct Generation, NFE=2)
11.1
FID· 2023-03-02
Consistency Models Code
#30CT (Direct Generation, NFE=1)
13
FID· 2023-03-02
Consistency Models Code
#31FM
14.45
FID· 2022-10-06
Flow Matching for Generative Modeling Code
#32CLR-GAN
20.27
FID
No paperCode
#33PGMGAN
21.73
FID· 2021-04-02
Partition-Guided GANs Code
#34GLIDE + CLIP + CLS + CLS-FREE
29.184
FID· 2022-10-20
Composing Ensembles of Pre-trained Models via Iterative Consensus
#35GLIDE + CLS-FREE
29.219
FID· 2022-10-20
Composing Ensembles of Pre-trained Models via Iterative Consensus
#36GLIDE + CLIP
30.462
FID· 2022-10-20
Composing Ensembles of Pre-trained Models via Iterative Consensus
#37GLIDE + CLS
30.871
FID· 2022-10-20
Composing Ensembles of Pre-trained Models via Iterative Consensus