10-shot image generation on ADE20K

Metric: GFLOPs (512 x 512) (higher is better)

LeaderboardDataset

Loading chart...

Results

Sort:

#	Model↕	GFLOPs (512 x 512)▼	Extra Data	Paper	Date↕	Code
1	ConvNeXt-XL++	3335	No	A ConvNet for the 2020s	2022-01-10	Code
2	ConvNeXt-L++	2458	No	A ConvNet for the 2020s	2022-01-10	Code
3	ConvNeXt-B++	1828	No	A ConvNet for the 2020s	2022-01-10	Code
4	GC ViT-B	1348	No	Global Context Vision Transformers	2022-06-20	Code
5	MogaNet-L (UperNet)	1176	No	MogaNet: Multi-order Gated Aggregation Network	2022-11-07	Code
6	ConvNeXt-B	1170	No	A ConvNet for the 2020s	2022-01-10	Code
7	GC ViT-S	1163	No	Global Context Vision Transformers	2022-06-20	Code
8	NAT-Base	1137	No	Neighborhood Attention Transformer	2022-04-14	Code
9	MogaNet-B (UperNet)	1050	No	MogaNet: Multi-order Gated Aggregation Network	2022-11-07	Code
10	ConvNeXt-S	1027	No	A ConvNet for the 2020s	2022-01-10	Code
11	NAT-Small	1010	No	Neighborhood Attention Transformer	2022-04-14	Code
12	GC ViT-T	947	No	Global Context Vision Transformers	2022-06-20	Code
13	MogaNet-S (UperNet)	946	No	MogaNet: Multi-order Gated Aggregation Network	2022-11-07	Code
14	ConvNeXt-T	939	No	A ConvNet for the 2020s	2022-01-10	Code
15	NAT-Tiny	934	No	Neighborhood Attention Transformer	2022-04-14	Code
16	NAT-Mini	900	No	Neighborhood Attention Transformer	2022-04-14	Code
17	EoMT (DINOv2-L, single-scale, 512x512)	721	No	Your ViT is Secretly an Image Segmentation Model	2025-03-24	Code
18	SegViT-v2 (BEiT-v2-Large)	637.9	No	SegViTv2: Exploring Efficient and Continual Sema...	2023-06-09	Code
19	MogaNet-S (Semantic FPN)	189	No	MogaNet: Multi-order Gated Aggregation Network	2022-11-07	Code
20	Light-Ham (VAN-Huge)	71.8	No	Is Attention Better Than Matrix Decomposition?	2021-09-09	Code
21	HRViT-b3 (SegFormer, SS)	67.9	No	Multi-Scale High-Resolution Vision Transformer f...	2021-11-01	Code
22	Light-Ham (VAN-Large)	55	No	Is Attention Better Than Matrix Decomposition?	2021-09-09	Code
23	Light-Ham (VAN-Base)	34.4	No	Is Attention Better Than Matrix Decomposition?	2021-09-09	Code
24	HRViT-b2 (SegFormer, SS)	28	No	Multi-Scale High-Resolution Vision Transformer f...	2021-11-01	Code
25	Light-Ham (VAN-Small, D=256)	15.8	No	Is Attention Better Than Matrix Decomposition?	2021-09-09	Code
26	HRViT-b1 (SegFormer, SS)	14.6	No	Multi-Scale High-Resolution Vision Transformer f...	2021-11-01	Code

#1ConvNeXt-XL++SOTA
3335
GFLOPs (512 x 512)· 2022-01-10
A ConvNet for the 2020s Code
#2ConvNeXt-L++
2458
GFLOPs (512 x 512)· 2022-01-10
A ConvNet for the 2020s Code
#3ConvNeXt-B++
1828
GFLOPs (512 x 512)· 2022-01-10
A ConvNet for the 2020s Code
#4GC ViT-B
1348
GFLOPs (512 x 512)· 2022-06-20
Global Context Vision Transformers Code
#5MogaNet-L (UperNet)
1176
GFLOPs (512 x 512)· 2022-11-07
MogaNet: Multi-order Gated Aggregation Network Code
#6ConvNeXt-B
1170
GFLOPs (512 x 512)· 2022-01-10
A ConvNet for the 2020s Code
#7GC ViT-S
1163
GFLOPs (512 x 512)· 2022-06-20
Global Context Vision Transformers Code
#8NAT-Base
1137
GFLOPs (512 x 512)· 2022-04-14
Neighborhood Attention Transformer Code
#9MogaNet-B (UperNet)
1050
GFLOPs (512 x 512)· 2022-11-07
MogaNet: Multi-order Gated Aggregation Network Code
#10ConvNeXt-S
1027
GFLOPs (512 x 512)· 2022-01-10
A ConvNet for the 2020s Code
#11NAT-Small
1010
GFLOPs (512 x 512)· 2022-04-14
Neighborhood Attention Transformer Code
#12GC ViT-T
947
GFLOPs (512 x 512)· 2022-06-20
Global Context Vision Transformers Code
#13MogaNet-S (UperNet)
946
GFLOPs (512 x 512)· 2022-11-07
MogaNet: Multi-order Gated Aggregation Network Code
#14ConvNeXt-T
939
GFLOPs (512 x 512)· 2022-01-10
A ConvNet for the 2020s Code
#15NAT-Tiny
934
GFLOPs (512 x 512)· 2022-04-14
Neighborhood Attention Transformer Code
#16NAT-Mini
900
GFLOPs (512 x 512)· 2022-04-14
Neighborhood Attention Transformer Code
#17EoMT (DINOv2-L, single-scale, 512x512)
721
GFLOPs (512 x 512)· 2025-03-24
Your ViT is Secretly an Image Segmentation Model Code
#18SegViT-v2 (BEiT-v2-Large)
637.9
GFLOPs (512 x 512)· 2023-06-09
SegViTv2: Exploring Efficient and Continual Semantic Segmentation with Plain Vision Transformers Code
#19MogaNet-S (Semantic FPN)
189
GFLOPs (512 x 512)· 2022-11-07
MogaNet: Multi-order Gated Aggregation Network Code
#20Light-Ham (VAN-Huge)SOTA
71.8
GFLOPs (512 x 512)· 2021-09-09
Is Attention Better Than Matrix Decomposition?Code
#21HRViT-b3 (SegFormer, SS)
67.9
GFLOPs (512 x 512)· 2021-11-01
Multi-Scale High-Resolution Vision Transformer for Semantic Segmentation Code
#22Light-Ham (VAN-Large)
55
GFLOPs (512 x 512)· 2021-09-09
Is Attention Better Than Matrix Decomposition?Code
#23Light-Ham (VAN-Base)
34.4
GFLOPs (512 x 512)· 2021-09-09
Is Attention Better Than Matrix Decomposition?Code
#24HRViT-b2 (SegFormer, SS)
28
GFLOPs (512 x 512)· 2021-11-01
Multi-Scale High-Resolution Vision Transformer for Semantic Segmentation Code
#25Light-Ham (VAN-Small, D=256)
15.8
GFLOPs (512 x 512)· 2021-09-09
Is Attention Better Than Matrix Decomposition?Code
#26HRViT-b1 (SegFormer, SS)
14.6
GFLOPs (512 x 512)· 2021-11-01
Multi-Scale High-Resolution Vision Transformer for Semantic Segmentation Code