Image Generation on ImageNet 256x256

Metric: FID (lower is better)

LeaderboardDataset

Loading chart...

Results

#	Model↕	FID▲	Extra Data	Paper	Date↕	Code
1	SiT-XL/2 + UCGM-S (E2E-VAE + 40 sampling steps + CFG)	1.06	No	Unified Continuous Generative Models	2025-05-12	Code
2	UCGM-XL/2 (VA-VAE + 30 sampling steps, without guidance)	1.21	No	Unified Continuous Generative Models	2025-05-12	Code
3	UCGM-XL/2 (E2E-VAE + 40 sampling steps, without guidance)	1.21	No	Unified Continuous Generative Models	2025-05-12	Code
4	EDM2-L + DDO (SD-VAE, 25 steps, DPM-Solver-v3)	1.21	No	Direct Discriminative Optimization: Your Likelih...	2025-03-03	Code
5	LightningDiT + UCGM-S (VA-VAE + 50 sampling steps + CFG)	1.21	No	Unified Continuous Generative Models	2025-05-12	Code
6	xAR-H	1.24	No	Beyond Next-Token: Next-X Prediction for Autoreg...	2025-02-27	Code
7	SiT-XL/2 + REPA-E	1.26	No	REPA-E: Unlocking VAE for End-to-End Tuning with...	2025-04-14	Code
8	DDT-XL/2(22en6de 675M + guidance interval )	1.26	No	DDT: Decoupled Diffusion Transformer	2025-04-08	Code
9	xAR-L	1.28	No	Beyond Next-Token: Next-X Prediction for Autoreg...	2025-02-27	Code
10	FACM (2-step)	1.32	No	Flow-Anchored Consistency Models	2025-07-04	Code
11	GMem (with the guidance interval)	1.32	No	Generative Modeling with Explicit Memory	2024-12-11	Code
12	SiT-XL/2 + MG	1.34	No	Diffusion Models without Classifier-free Guidance	2025-02-17	Code
13	AliTok-XL, autoregressive, 662M	1.35	No	AliTok: Towards Sequence Modeling Alignment betw...	2025-06-05	Code
14	LightningDiT + VA-VAE (with the guidance interval)	1.35	No	Reconstruction vs. Generation: Taming Optimizati...	2025-01-02	Code
15	SiD2	1.38	No	Simpler Diffusion (SiD2): 1.5 FID on ImageNet512...	2024-10-25	-
16	SiT↓-XL/2+U-REPA (with the guidance interval)	1.41	No	U-REPA: Aligning Diffusion U-Nets to ViTs	2025-03-24	Code
17	AliTok-XL, autoregressive, 318M	1.42	No	AliTok: Towards Sequence Modeling Alignment betw...	2025-06-05	Code
18	SiT-XL/2 + REPA (with the guidance interval)	1.42	No	Representation Alignment for Generation: Trainin...	2024-10-09	Code
19	RAR-XXL, autoregressive	1.48	No	Randomized Autoregressive Visual Generation	2024-11-01	Code
20	RAR-XL, autoregressive	1.5	No	Randomized Autoregressive Visual Generation	2024-11-01	Code
21	MaskBit	1.52	No	MaskBit: Embedding-free Image Generation via Bit...	2024-09-24	Code
22	GMem (w/o guidance)	1.53	No	Generative Modeling with Explicit Memory	2024-12-11	Code
23	ELM	1.54	No	Elucidating the design space of language models ...	2024-10-21	Code
24	MAR-H, Diff Loss	1.55	No	Autoregressive Image Generation without Vector Q...	2024-06-17	Code
25	PaGoDA	1.56	No	PaGoDA: Progressive Growing of a One-Step Genera...	2024-05-23	Code
26	ViT-XL/2 with limited Interval Guidance	1.57	No	Efficient Diffusion Training via Min-SNR Weighti...	2023-03-16	Code
27	MDTv2	1.58	No	MDTv2: Masked Diffusion Transformer is a Strong ...	2023-03-25	Code
28	SiT-XL + SRA	1.58	No	No Other Representation Component Is Needed: Dif...	2025-05-05	Code
29	RobustTok-L	1.6	No	Robust Latent Matters: Boosting Image Generation...	2025-03-11	Code
30	DiMR-G/2R	1.63	No	Alleviating Distortion in Image Generation via M...	2024-06-13	Code
31	FlowAR	1.65	No	FlowAR: Scale-wise Autoregressive Image Generati...	2024-12-19	Code
32	FACM (1-step)	1.7	No	Flow-Anchored Consistency Models	2025-07-04	Code
33	DiT-XL/2 with CADS	1.7	No	CADS: Unleashing the Diversity of Diffusion Mode...	2023-10-26	-
34	DiMR-XL/2R	1.7	No	Alleviating Distortion in Image Generation via M...	2024-06-13	Code
35	RAR-L, autoregressive	1.7	No	Randomized Autoregressive Visual Generation	2024-11-01	Code
36	DiffiT	1.73	No	DiffiT: Diffusion Vision Transformers for Image ...	2023-12-04	Code
37	VAR (Visual Autoregressive)	1.73	No	Visual Autoregressive Modeling: Scalable Image G...	2024-04-03	Code
38	MAGVIT-v2	1.78	No	Language Model Beats Diffusion -- Tokenizer is K...	2023-10-09	Code
39	MAR-L, Diff Loss	1.78	No	Autoregressive Image Generation without Vector Q...	2024-06-17	Code
40	MDT	1.79	No	MDTv2: Masked Diffusion Transformer is a Strong ...	2023-03-25	Code
41	Discriminator Guidance	1.83	No	Refining Generative Process with Discriminator G...	2022-11-28	Code
42	DoD-XL	1.83	No	Diffusion Models Need Visual Priors for Image Ge...	2024-10-11	-
43	RobustTok-B	1.83	No	Robust Latent Matters: Boosting Image Generation...	2025-03-11	Code
44	ARPG-XXL	1.94	No	Autoregressive Image Generation with Randomized ...	2025-03-13	Code
45	RAR-B, autoregressive	1.95	No	Randomized Autoregressive Visual Generation	2024-11-01	Code
46	TiTok-S-128	1.97	No	An Image is Worth 32 Tokens for Reconstruction a...	2024-06-11	Code
47	PixelFlow	1.98	No	PixelFlow: Pixel-Space Generative Models with Flow	2025-04-10	Code
48	RDM	1.99	No	Relay Diffusion: Unifying diffusion process acro...	2023-09-04	Code
49	FasterDiT-XL/2	2.03	No	FasterDiT: Towards Faster Diffusion Transformers...	2024-10-14	Code
50	LEGO-XL	2.05	No	Learning Stackable and Skippable LEGO Bricks for...	2023-10-10	Code
51	ARPG-XL	2.1	No	Autoregressive Image Generation with Randomized ...	2025-03-13	Code
52	StyleSAN-XL	2.14	No	SAN: Inducing Metrizability of GAN with Discrimi...	2023-01-30	Code
53	LlamaGen	2.18	No	Autoregressive Model Beats Diffusion: Llama for ...	2024-06-10	Code
54	DiT-XL/2	2.27	No	Scalable Diffusion Models with Transformers	2022-12-19	Code
55	StyleGAN-XL	2.3	No	StyleGAN-XL: Scaling StyleGAN to Large Diverse D...	2022-02-01	Code
56	MAR-B, Diff Loss	2.31	No	Autoregressive Image Generation without Vector Q...	2024-06-17	Code
57	Open-MAGVIT2-XL	2.33	No	Open-MAGVIT2: An Open-Source Project Toward Demo...	2024-09-06	Code
58	ACDiT	2.37	No	ACDiT: Interpolating Autoregressive Conditional ...	2024-12-10	Code
59	ARPG-L	2.44	No	Autoregressive Image Generation with Randomized ...	2025-03-13	Code
60	TiTok-B-64	2.48	No	An Image is Worth 32 Tokens for Reconstruction a...	2024-06-11	Code
61	GIVT-Causal-L+A	2.59	No	GIVT: Generative Infinite-Vocabulary Transformers	2023-12-04	Code
62	Patch Diffusion	2.74	No	-	-	-
63	TiTok-B-32	2.77	No	An Image is Worth 32 Tokens for Reconstruction a...	2024-06-11	Code
64	DoD-B	2.79	No	Diffusion Models Need Visual Priors for Image Ge...	2024-10-11	-
65	Poly-INR	2.86	No	Polynomial Implicit Neural Representations For L...	2023-03-20	Code
66	MGVQ	3.02	No	MGVQ: Could VQ-VAE Beat VAE? A Generalizable Tok...	2025-07-10	Code
67	ADM-G++ (FID)	3.18	No	Refining Generative Process with Discriminator G...	2022-11-28	Code
68	DiGIT-0.7B	3.39	No	Stabilize the Latent Space for Image Autoregress...	2024-10-16	Code
69	DiGIT	3.39	No	Stabilize the Latent Space for Image Autoregress...	2024-10-16	Code
70	Contextual RQ-Transformer	3.41	No	Draft-and-Revise: Effective Image Generation wit...	2022-06-09	-
71	GigaGAN	3.45	No	Scaling up GANs for Text-to-Image Synthesis	2023-03-09	Code
72	RCG-L (w/o guidance)	3.49	No	Return of Unconditional Generation: A Self-super...	2023-12-06	Code
73	BIGRoC-gt (Guided-Diffusion)	3.63	No	BIGRoC: Boosting Image Generation via a Robust C...	2021-08-08	Code
74	MAGVIT-v2 (w/o guidance)	3.65	No	Language Model Beats Diffusion -- Tokenizer is K...	2023-10-09	Code
75	BIGRoC-pl (Guided-Diffusion)	3.69	No	BIGRoC: Boosting Image Generation via a Robust C...	2021-08-08	Code
76	simple diffusion (U-Net)	3.71	No	Simple diffusion: End-to-end diffusion for high ...	2023-01-26	Code
77	simple diffusion (U-ViT, L)	3.75	No	Simple diffusion: End-to-end diffusion for high ...	2023-01-26	Code
78	RQ-Transformer	3.83	No	Autoregressive Image Generation using Residual Q...	2022-03-03	Code
79	ADM-G, ADM-U	3.94	No	Diffusion Models Beat GANs on Image Synthesis	2021-05-11	Code
80	ADM-G + EDS (ED-DPM, classifier_scale=0.75)	3.96	No	Entropy-driven Sampling and Training Scheme for ...	2022-06-23	Code
81	MaskGIT (a=0.05)	4.02	No	MaskGIT: Masked Generative Image Transformer	2022-02-08	Code
82	ADM-G + EDS + ECT (ED-DPM, classifier_scale=1.0)	4.09	No	Entropy-driven Sampling and Training Scheme for ...	2022-06-23	Code
83	ADM-G + EDS + ECT (ED-DPM, classifier_scale=1.0)	4.09	No	Entropy-driven Sampling and Training Scheme for ...	2022-06-23	Code
84	LDM	4.29	No	-	-	-
85	ADM-G++ (Recall)	4.45	No	Refining Generative Process with Discriminator G...	2022-11-28	Code
86	LFM	4.46	No	Flow Matching in Latent Space	2023-07-17	Code
87	RIN	4.51	No	Scalable Adaptive Computation for Iterative Gene...	2022-12-22	Code
88	ADM-G	4.59	No	Diffusion Models Beat GANs on Image Synthesis	2021-05-11	Code
89	ADM-G	4.59	No	Diffusion Models Beat GANs on Image Synthesis	2021-05-11	Code
90	CDM	4.88	No	Cascaded Diffusion Models for High Fidelity Imag...	2021-05-30	-
91	VQGAN+Transformer (k=600, p=1.0, a=0.05)	5.2	No	Taming Transformers for High-Resolution Image Sy...	2020-12-17	Code
92	MaskGIT	6.18	No	MaskGIT: Masked Generative Image Transformer	2022-02-08	Code
93	VQGAN+Transformer (k=mixed, p=1.0, a=0.005)	6.59	No	Taming Transformers for High-Resolution Image Sy...	2020-12-17	Code
94	Polarity-BigGAN	6.82	No	Polarity Sampling: Quality and Diversity Control...	2022-03-03	Code
95	BigGAN-deep	8.1	No	Large Scale GAN Training for High Fidelity Natur...	2018-09-28	Code
96	BigGAN+ [Brock et al.] (chx96)	8.1	No	Instance-Conditioned GAN	2021-09-10	Code
97	ADM	11.84	No	-	-	-
98	Improved DDPM	12.3	No	Improved Denoising Diffusion Probabilistic Models	2021-02-18	Code