MambaVision: A Hybrid Mamba-Transformer Vision Backbone

Open paper
Benchmark Model Rank Results
image-classification-on-imagenetMambaVision-L3#58Top 1 Accuracy: 88.1%GFLOPs: 489.1
image-classification-on-imagenetMambaVision-L#263Top 1 Accuracy: 85%Number of params: 227.9MGFLOPs: 34.9
image-classification-on-imagenetMambaVision-B#335Top 1 Accuracy: 84.2%Number of params: 97.7MGFLOPs: 15
image-classification-on-imagenetMambaVision-S#434Top 1 Accuracy: 83.3%Number of params: 50.1MGFLOPs: 7.5
image-classification-on-imagenetMambaVision-T2#500Top 1 Accuracy: 82.7%Number of params: 35.1MGFLOPs: 5.1
image-classification-on-imagenetMambaVision-T#538Top 1 Accuracy: 82.3%Number of params: 31.8MGFLOPs: 4.4
image-classification-on-imagenetMambaVision-L2#985Number of params: 241.5M

Rank counts only results with a code link.