A2Mamba: Attention-augmented State Space Models for Visual Recognition

Open paper
Benchmark Model Rank Results
semantic-segmentation-on-ade20kA2Mamba-L#71Validation mIoU: 54.1Params (M): 126GFLOPs (512 x 512): 1237
semantic-segmentation-on-ade20kA2Mamba-B#75Validation mIoU: 53.9Params (M): 80GFLOPs (512 x 512): 1090
semantic-segmentation-on-ade20kA2Mamba-S#97Validation mIoU: 52.0Params (M): 60GFLOPs (512 x 512): 959
image-classification-on-imagenetA2Mamba-S#322Top 1 Accuracy: 84.7%GFLOPs: 5.4
image-classification-on-imagenetA2Mamba-T#514Top 1 Accuracy: 83.0%GFLOPs: 2.7
image-classification-on-imagenetA2Mamba-N#851Top 1 Accuracy: 78.7%GFLOPs: 0.8

Rank counts only results with a code link.