| visual-grounding-on-refcoco-plus-test-b | Florence-2-large-ft | #1 | Accuracy (%): 92.0 |
| visual-grounding-on-refcoco-plus-testa | Florence-2-large-ft | #1 | Accuracy (%): 95.3 |
| visual-grounding-on-refcoco-plus-val | Florence-2-large-ft | #1 | Accuracy (%): 93.4 |
| image-captioning-on-nocaps-val-overall | Florence-2-L | #2 | CIDEr: 124.9 |
| visual-question-answering-on-vizwiz-2020-vqa | Florence-2-L | #2 | overall: 72.6 |
| image-captioning-on-textcaps | Florence-2-L | #3 | CIDEr: 151.1 |
| visual-question-answering-on-vqa-v2-test-dev | Florence-2-L | #9 | Accuracy: 81.7 |
| instance-segmentation-on-coco-minival | DaViT-B + Mask R-CNN (Florence-2 pretrain) | #52 | mask AP: 46.4 |
| semantic-segmentation-on-ade20k | DaViT-B + UperNet (Florence-2 pretrain) | #56 | Validation mIoU: 54.9 |